# GLLmolfar documentation > What every screen of GLLmolfar, the desktop crawler, shows: each tab, block, figure, report and setting, in the words the app prints. Describes GLLmolfar 0.9.22. Every page is also served as Markdown at the same address with .md added; https://www.getlooploop.com/gllmolfar/docs/llms-full.txt holds all of them in one file. GLLmolfar is a desktop crawler for Windows and Mac. It crawls a site on the user's computer, embeds page text with a local model, and reports technical SEO, content and AI-visibility findings. AI agents connect to it over MCP. ## Start here - [The app at a glance](https://www.getlooploop.com/gllmolfar/docs/what-gllmolfar-shows.md): The GLLmolfar window: the menu bar, the toolbar, the rail of sections and its panel, Ctrl K, the status bar, Simple and Pro, and the Hybrid and AI focus. ## Crawl and overview - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. ## Reports - [Reports](https://www.getlooploop.com/gllmolfar/docs/reports.md): The Reports tab: eleven reports in four directions, each answering one question about the open crawl, with a language filter and HTML or PDF export. ## AI and agents - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. - [Agent · test](https://www.getlooploop.com/gllmolfar/docs/agent-test.md): The Agent · test screen: your own models plan tasks from the fix list, rewrite pages, write social posts and review risky pages; GLLmolfar measures the results. - [Reverse prompts](https://www.getlooploop.com/gllmolfar/docs/reverse-prompts.md): Site › Reverse prompts: GLLmolfar reads a page into blocks, your AI agent describes what each does, and every block gets a reusable prompt and an example. - [MCP tools for agents](https://www.getlooploop.com/gllmolfar/docs/mcp-tools.md): The MCP server inside GLLmolfar: how to connect an AI agent, and every tool it can call to crawl, read fixes, audit pages, search by meaning and save reports. ## Projects and history - [Projects](https://www.getlooploop.com/gllmolfar/docs/projects.md): The Projects tab: one card per site crawled on this computer, with its health, how the last crawl changed, its scans over time, topics and domain. - [Schedule](https://www.getlooploop.com/gllmolfar/docs/schedule.md): The Schedule tab: a month calendar of planned and past crawls, the list of schedules, and the dialog that plans a crawl once or on repeat. - [History](https://www.getlooploop.com/gllmolfar/docs/history.md): The History tab: every saved crawl of a site, a trend of one number across them, and a comparison of two crawls page by page and check by check. ## Checks - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. - [Fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list.md): The Fix list tab: what to fix in the crawled site, numbered, most important first, each with why it matters, how to fix it and the URLs. - [Batch fixes](https://www.getlooploop.com/gllmolfar/docs/batch-fixes.md): The Batch fixes tab: problems grouped by the one place that fixes many pages at once — the server, the shared layout, a template or a topic's texts. ## Meaning - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map.md): The Topic map tab: every embedded page as a dot, placed by meaning and coloured by topic, status, indexability or depth, with a card for each page. - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization.md): The Cannibalization tab: groups of pages that say almost the same thing, the page to keep, the pages side by side and their text passage by passage. - [Link opportunities](https://www.getlooploop.com/gllmolfar/docs/link-opportunities.md): The Link opportunities tab: weakly linked pages, the pages closest to them in meaning that do not link to them yet, and an anchor for each link. - [Semantic search](https://www.getlooploop.com/gllmolfar/docs/semantic-search.md): The Semantic search tab: type a question and get the crawled pages closest to it in meaning, with the sections of each page that match best. ## Site - [All pages (Internal)](https://www.getlooploop.com/gllmolfar/docs/all-pages.md): The All pages tab, called Internal in Pro mode: every crawled URL in one table, with filters, search, column sets, an overview panel and a page inspector. - [Structure](https://www.getlooploop.com/gllmolfar/docs/structure.md): The Structure tab: the crawled site drawn as a crawl tree, a directory tree or a force-directed diagram, coloured by topic, status, depth, speed or a check. - [URL audit](https://www.getlooploop.com/gllmolfar/docs/url-audit.md): Site › URL audit: one page checked without a crawl — status, speed, JavaScript, page map, failed checks, accessibility, robots.txt, headers and screenshots. - [Heat map](https://www.getlooploop.com/gllmolfar/docs/heat-map.md): Site › Heat map: the crawl's folders as tiles sized by pages and coloured by pages without problems, problems per page, Google clicks or lost clicks. ## Search Console - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Pages](https://www.getlooploop.com/gllmolfar/docs/gsc-pages.md): GSC › Pages: each page's Google clicks, impressions, CTR and position, its top query, its state in the crawl, and crawled pages Google never showed. - [Search Console: Positions](https://www.getlooploop.com/gllmolfar/docs/gsc-positions.md): GSC › Positions: the period's queries in five bands of average position, and the queries on positions 4 to 20 where a small gain moves clicks most. - [Search Console: Indexing](https://www.getlooploop.com/gllmolfar/docs/gsc-indexing.md): GSC › Indexing: Google's URL Inspection answer page by page: in the index or not and why, last crawl, chosen canonical, rich results, kept on this computer. - [Search Console: Sitemaps](https://www.getlooploop.com/gllmolfar/docs/gsc-sitemaps.md): GSC › Sitemaps: the sitemaps submitted to Search Console, their URLs, errors and warnings, when Google last read them, beside the URLs the crawl found. - [Search Console: Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks.md): GSC › Lost clicks: pages that lost Google clicks over the last 28 days, the reason the numbers point at, the lost queries, and what the crawl found there. - [Search Console: Cannibalization](https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization.md): GSC › Cannibalization: queries for which Google showed two or more of your pages, with each page's clicks, share of impressions and position. - [Search Console: Query clusters](https://www.getlooploop.com/gllmolfar/docs/gsc-query-clusters.md): GSC › Query clusters: up to 1000 Search Console queries grouped by meaning on this computer, with each cluster's clicks, position, pages and queries. - [Search Console: Visualisation](https://www.getlooploop.com/gllmolfar/docs/gsc-visualisation.md): GSC › Visualisation: the period's queries drawn: words as a treemap, spread by position, CTR and impressions, opportunities, click concentration, brand. ## Free tools - [Free tools](https://www.getlooploop.com/gllmolfar/docs/free-tools.md): The Free tools section: Vectorscope, Voicescope, Brandscope, Reviewscope and PatternScope from getlooploop.com, run inside GLLmolfar on its own allowance. ## Settings - [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general.md): Settings › General: the interface language, term explanations, the interface mode, the default crawl size and mode, updates, statistics, the tray, MCP, sound. - [Settings › Crawling](https://www.getlooploop.com/gllmolfar/docs/settings-crawling.md): Settings › Crawling: request speed, connections, the user agent, the timeout, robots.txt, a pause at a CAPTCHA, JavaScript rendering and service pages. - [Settings › Crawl scope](https://www.getlooploop.com/gllmolfar/docs/settings-scope.md): Settings › Crawl scope: which URLs a crawl may visit — include and exclude patterns, depth, URLs with parameters, subdomains, nofollow and external links. - [Settings › Requests](https://www.getlooploop.com/gllmolfar/docs/settings-requests.md): Settings › Requests: extra HTTP headers sent with every page request, and a user name and password for a staging site behind Basic authentication. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. - [Settings › AI models](https://www.getlooploop.com/gllmolfar/docs/settings-ai-models.md): Settings › AI models: connect your own language models, on this computer, on your network or in a cloud, and choose which model does which analysis. - [Settings › Planned crawls](https://www.getlooploop.com/gllmolfar/docs/settings-planned.md): Settings › Planned crawls: what happens to a run the computer was off for, starting with the system in the tray, and notifications for planned crawls. - [Settings › Storage](https://www.getlooploop.com/gllmolfar/docs/storage.md): Settings › Storage: how much disk the saved crawls, the embedding model and downloaded updates take, per project, and the actions that free it. ## Optional - [GLLmolfar](https://www.getlooploop.com/gllmolfar): what the app is, and how to request it --- # The app at a glance > One window. A menu bar along the top, a toolbar that starts a crawl, a rail on the left — a strip of sections and a panel with the open section's views — the open view in the middle, and a status bar at the bottom that says what the app is doing. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/what-gllmolfar-shows - **Where:** The main window - **Mode:** Simple and Pro - **Built from:** The open crawl - **Changes:** With every crawl you start or open ## On the screen The window in Simple mode with a saved crawl open on the [Overview](https://www.getlooploop.com/gllmolfar/docs/overview). ![The GLLmolfar window with a saved crawl of getlooploop.com open on the Overview](https://www.getlooploop.com/images/gllmolfar/docs/overview.webp) Numbered parts of the screenshot: 1. **Menu bar.** **File**, **Edit**, **View**, **Crawl**, **Bulk export**, **Reports**, **Visualisations** and **Help**, after the app's name and version. On the right: **Feedback**, **Sign in with Google** or the signed-in account, and the gear that opens Settings. 2. **Toolbar.** **URL** and **Max URLs** (and **Mode** in Pro), **Start**, **Clear**, the **Embeddings** box for this crawl, and **Export** for the table on screen. 3. **Section strip.** Overview, Reports, Checks, Meaning, Site, GSC, AI, Free tools and Projects. Pressing one opens its views in the panel beside it. 4. **Panel.** The open crawl at the top, **Go to a screen…**, then the views of the open section. Here the section is Overview, with its one view. 5. **Hide and Ctrl K.** **Hide** folds the panel (**Show** brings it back; **Ctrl B** does both). **Ctrl K** opens the palette of every screen. 6. **Simple / Pro, Hybrid / AI.** Two switches at the foot of the panel: how much the app shows, and whom the checks are rated for. 7. **Status bar.** The crawl's state, the crawl counter and its bar, the fragments counter for embeddings, and under them **RAM**, **CPU**, **GPU**, **Session** time and the version. ## The menu bar | Menu | What is in it | | --- | --- | | File | New crawl, Open saved crawl…, Export this table…, Report (PDF / HTML)…, Settings…, Exit. | | Edit | Undo, Redo, Cut, Copy, Paste, Select all, Find in table. | | View | The main views (Overview, Internal, Issues, Fix list, Topics, Topic map, Cannibalization, Link opportunities, Semantic search, History, Projects), Columns…, Folder tree, Overview panel, Simple mode or Pro mode, zoom and Full screen. | | Crawl | Start crawl, Pause crawl or Resume crawl, Stop crawl, Clear, Crawl a list of URLs…, Add URLs to this crawl…, Quick URL audit, Crawl configuration…, Custom extraction & search…, the Embeddings switch, Open data folder. | | Bulk export | All internal links, All external links, Broken links (4xx, 5xx) with sources, Links that point to redirects, Images without alt, All images. | | Reports | Crawl overview, Issues overview, Fix list, Batch fixes, Report (PDF / HTML)…, Redirects, Canonicals, Hreflang, Insecure content (http on https), SERP summary (titles, descriptions, pixels), Orphan pages, Generate XML sitemap…. | | Visualisations | Crawl tree, Directory tree, Force-directed, Topic map, Site depth chart, About the page (selected URL), Topics, Cannibalization, Link opportunities, Semantic search. | | Help | User guide, Keyboard shortcuts, Show tips again, Relax, GetLoopLoop website, Open data folder, Check for updates…, Connect an AI agent…, Agent skills…, Send feedback or report a bug…, About GLLmolfar. | Items that need a crawl stay grey until one is open. **User guide** opens these pages. **Send feedback or report a bug…** can attach the app's diagnostic log to a bug report; the box **Attach the diagnostic log** decides. ## The toolbar | Control | What it does | | --- | --- | | URL | The address to crawl. Earlier start addresses are offered as you type. | | Mode | Spider + sitemap, Spider only, Sitemap only, or List of URLs. Pro mode only; Simple mode uses the default mode from [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general). | | Max URLs | Where the crawl stops. | | Start | Starts the crawl. While it runs, the button reads **Stop**, and **Pause** appears beside it; a paused crawl shows **Resume**. | | Add URLs | During a crawl: adds addresses to its queue. | | Clear | Shown when nothing is crawling: empties the window for a new crawl. Saved crawls stay saved. | | Embeddings | Whether this crawl builds embeddings for topics, cannibalization and semantic search. Its starting state comes from [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings). | | Export | Exports the table on screen. | ## The left rail Two columns: a narrow strip of sections, and beside it a panel with the views of the section that is open. Pressing a section opens its panel; pressing the open section again folds the panel away and gives the view the room. **Overview** has one view and opens it at once. | Section | Views in its panel | | --- | --- | | Overview | [Overview](https://www.getlooploop.com/gllmolfar/docs/overview) | | Reports | The eleven [reports](https://www.getlooploop.com/gllmolfar/docs/reports), under four headings: Technical SEO, Content and optimisation, AI visibility, Demand and progress. | | Checks | [Issues](https://www.getlooploop.com/gllmolfar/docs/issues), [Fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list), [Batch fixes](https://www.getlooploop.com/gllmolfar/docs/batch-fixes) | | Meaning | [Topics](https://www.getlooploop.com/gllmolfar/docs/topics), [Topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map), [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization), [Link opportunities](https://www.getlooploop.com/gllmolfar/docs/link-opportunities), [Semantic search](https://www.getlooploop.com/gllmolfar/docs/semantic-search) | | Site | [Internal](https://www.getlooploop.com/gllmolfar/docs/all-pages) (called **All pages** in Simple mode), [Structure](https://www.getlooploop.com/gllmolfar/docs/structure), [URL audit](https://www.getlooploop.com/gllmolfar/docs/url-audit), [Heat map](https://www.getlooploop.com/gllmolfar/docs/heat-map), [Reverse prompts](https://www.getlooploop.com/gllmolfar/docs/reverse-prompts) | | GSC | The nine Search Console reports: [Performance](https://www.getlooploop.com/gllmolfar/docs/search-console), [Pages](https://www.getlooploop.com/gllmolfar/docs/gsc-pages), [Positions](https://www.getlooploop.com/gllmolfar/docs/gsc-positions), [Indexing](https://www.getlooploop.com/gllmolfar/docs/gsc-indexing), [Sitemaps](https://www.getlooploop.com/gllmolfar/docs/gsc-sitemaps), [Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks), [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization), [Query clusters](https://www.getlooploop.com/gllmolfar/docs/gsc-query-clusters), [Visualisation](https://www.getlooploop.com/gllmolfar/docs/gsc-visualisation) | | AI | [AI](https://www.getlooploop.com/gllmolfar/docs/ai) and [Agent · test](https://www.getlooploop.com/gllmolfar/docs/agent-test) | | Free tools | Vectorscope, Voicescope, Brandscope, Reviewscope and PatternScope. See [Free tools](https://www.getlooploop.com/gllmolfar/docs/free-tools). | | Projects | [Projects](https://www.getlooploop.com/gllmolfar/docs/projects), [Schedule](https://www.getlooploop.com/gllmolfar/docs/schedule) and [History](https://www.getlooploop.com/gllmolfar/docs/history) | | Part | What it shows or does | | --- | --- | | GL | The top of the strip opens the Overview. While a crawl runs, a ring around it fills with the crawl's progress. | | Red dot | On **Checks**, when Issues has checks to count. | | Open crawl | At the top of the panel: the site, its URL count and date, and its health score. Pressing it lists the saved crawls. | | Go to a screen… | Opens the palette, as **Ctrl K** does. The last three screens opened sit under it, one click away. | | Panel heading | The section's name, and the panel icon that folds the panel. | | Hide / Show | At the foot of the strip: folds the panel and brings it back. **Ctrl B** does the same. | | Ctrl K | Under Hide: opens the palette. | The app remembers on this computer whether the panel is folded and which Reports headings are folded. A lock on a view means it is not open on this account; pressing it says what opens it. ## Go to a screen: Ctrl K Every screen by its name, in one list. Type part of a name or of a section, move with the arrows, open with Enter; Escape closes it. A report or a Search Console page opens exactly as its button in the panel does, and a locked view stays locked. ![The Ctrl K palette of GLLmolfar listing the screens whose names contain se](https://www.getlooploop.com/images/gllmolfar/docs/command-palette.webp) Numbered parts of the screenshot: 1. **Go to a screen….** The field at the top of the panel. Pressing it opens the palette, as **Ctrl K** does anywhere in the window. 2. **The field.** **Go to… a report, a tab, a check**. Here it holds **se**. 3. **Screens.** Each screen whose name or section contains the text. The one chosen is highlighted; Enter opens it. 4. **Section.** Where the screen lives on the rail: Reports, Meaning, Site, Search Console. 5. **Ctrl K on the strip.** The same palette from the foot of the strip. - **No match** — the app says: "Nothing by that name" No screen's name or section contains the text. ## Simple and Pro, Hybrid and AI Two switches sit at the foot of the panel. Simple and Pro can also be changed in **View** and in [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general). | Switch | Choice | What changes | | --- | --- | --- | | Simple / Pro | Simple | What to do and why. The panels leave out Issues, Cannibalization, Link opportunities and Semantic search, the page table is called **All pages**, and the toolbar has no **Mode**. | | Simple / Pro | Pro | Every table and filter: every view, the crawl tabs above the table, **Compare with**, column search, **Tree**, **Columns** and the **Overview panel**. | | Hybrid / AI | Hybrid | Search engines and AI assistants together. | | Hybrid / AI | AI | Only AI crawlers, assistants and agents: how they fetch, read and quote the site. Checks are rated for AI readers, and the score is AI readiness. | ## Waits, terms and tables | What | How it looks | | --- | --- | | A wait | One bar, the same on every screen: what is happening, the seconds so far (**N s**) and, once this kind of wait has been seen on this computer, **usually about** its usual length. When the work reports how far it is, the bar is that; otherwise it is the share of the usual time, held short of the end until the work is done. | | A technical term | nosnippet, X-Robots-Tag, hreflang, canonical, embeddings and the like have a dotted underline. Hovering or focusing one shows a sentence that explains it and **More in the GetLoopLoop glossary**. [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general) switches it off: **Explain technical terms (nosnippet, X-Robots-Tag…) on hover**. | | A table on a narrow window | No column is hidden and nothing scrolls sideways. First the text wraps and an address shrinks to one line with … (the full address is its tooltip); if that is not enough, each row becomes a small card with every cell under its column's name. The page table in [Internal](https://www.getlooploop.com/gllmolfar/docs/all-pages) keeps its own scrolling. | ## The status bar | Part | What it shows | | --- | --- | | State | A coloured dot and the state, with the site and date for a saved crawl, or a message about the crawl. | | Crawl | URLs crawled out of the crawl's ceiling, then what is in the queue, the speed, pages rendered with JavaScript, pages skipped by scope or met with a CAPTCHA, and external links checked. A bar follows. | | Fragments | The page sections turned into embeddings, with what is waiting, the speed and the time left; **saved with the crawl** when they are done. | | RAM | Memory used by GLLmolfar. The tooltip adds the embedding model's share and the computer's total. | | CPU | Share of all processor cores used by GLLmolfar. | | GPU | Load of the app's GPU process. | | Session | Time since GLLmolfar started. | | Version | The installed version; the tooltip adds when it was built. | RAM and CPU turn amber, then red, as they climb. ## States and messages - **Ready** — the app says: "Ready" Nothing is crawling and no crawl is open. - **Crawling** — the app says: "Crawling" A crawl is running. The counter and the queue move. - **Paused** — the app says: "Paused" The queue is kept; **Resume** continues it. - **Done** — the app says: "Crawl complete · working on the fragments" The pages are in; embeddings are still being made. - **Saved** — the app says: "Saved crawl" A saved crawl is open, with its site and date. - **Stopped** — the app says: "Stopped" Stopped by hand. The pages crawled so far are kept. - **Failed** — the app says: "Crawl failed" The crawl could not go on. The message beside it says why. - **AI focus** — the app says: "AI mode: checks are rated for AI readers, the score is AI readiness." After switching to **AI**. - **Hybrid focus** — the app says: "Hybrid mode: search engines and AI together." After switching to **Hybrid**. ## Related pages - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [All pages (Internal)](https://www.getlooploop.com/gllmolfar/docs/all-pages.md): The All pages tab, called Internal in Pro mode: every crawled URL in one table, with filters, search, column sets, an overview panel and a page inspector. - [Reports](https://www.getlooploop.com/gllmolfar/docs/reports.md): The Reports tab: eleven reports in four directions, each answering one question about the open crawl, with a language filter and HTML or PDF export. - [Free tools](https://www.getlooploop.com/gllmolfar/docs/free-tools.md): The Free tools section: Vectorscope, Voicescope, Brandscope, Reviewscope and PatternScope from getlooploop.com, run inside GLLmolfar on its own allowance. - [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general.md): Settings › General: the interface language, term explanations, the interface mode, the default crawl size and mode, updates, statistics, the tray, MCP, sound. - [MCP tools for agents](https://www.getlooploop.com/gllmolfar/docs/mcp-tools.md): The MCP server inside GLLmolfar: how to connect an AI agent, and every tool it can call to crawl, read fixes, audit pages, search by meaning and save reports. --- # The crawl screen > While a crawl runs, the Overview tab shows the run itself: how far it is, what each stage is doing and the last pages read. It stays until the fragments and the topics are done, then the Overview takes its place. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/crawl-screen - **Where:** Overview, during a run - **Mode:** Simple and Pro - **Built from:** The crawl in progress - **Changes:** Every second ## On the screen ![The crawl screen of GLLmolfar during a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/crawl-screen.webp) Numbered parts of the screenshot: 1. **Stages.** Crawl, Fragments, Topics and cannibalisation, Done. The stage in progress is marked; a finished one gets a tick. Done shows the forecast end time. 2. **Views.** **Overview**, **Neuro-pipeline** and **Terminal** show the same run three ways. **Compact** keeps each card to its numbers and the load chart as a small strip. 3. **Crawl card.** Pages read out of the target, the rate in pages a second, addresses found against the limit, and the time since the start. 4. **Fragments card.** Fragments embedded out of those found, fragments a second, the queue, and when the fragments end after the crawl. Two bars: first fragments, which give the topics, and more fragments, which refine them. 5. **Rules of the run.** Who the site sees (the user agent, to copy for a log search, and this computer's address), what robots.txt allows, and in Balanced mode the next checkpoint. 6. **Page checks.** Critical, warnings and info found so far, with the most frequent checks under each. The counts grow as the crawl reads. 7. **The whole run.** One chart of pages read and fragments waiting, with the forecast at the current pace. Below it: every fragment as a cell, the last pages read, and the load by stage. ## The four stages | Stage | While it runs | When it is done | | --- | --- | --- | | Crawl | N / target · percent | N pages · time taken | | Fragments | alongside · percent of those found, then embedding · percent | N fragments | | Topics and cannibalisation | after the fragments, then building the topics… | N topics · N pairs | | Done | ≈ the forecast end time | a tick | Fragments are the pieces of page text the model on this computer turns into vectors. Topics and cannibalisation pairs are built from them, so the third stage waits for the second. ## Cards and charts | Block | What it shows | | --- | --- | | Crawling the site | Pages read **of N pages**, the pages left, the rate, **Addresses found** with the limit, and the time since the start. | | Fragments | Fragments embedded **of N fragments**, fragments a second, and the queue: growing when pages come faster than the model reads them, falling, or empty. | | robots.txt | Whose rules the crawl follows (the site's, your own from Settings, or none), the pace, the number of streams, closed paths, sitemaps, and the AI bots the site shuts out. | | Balanced | In Balanced mode: the next checkpoint, or the one under way, where the crawl waits for the embeddings and saves the topics so far. | | Page checks | Findings at three levels (critical, warnings, info) filling as the crawl reads, the most frequent checks under each. When the crawl ends, a check opens its pages. | | The whole run | One chart: pages read, fragments waiting, the forecast at the current pace, and the moment the crawl completed. | | Fragments mosaic | Each fragment as a cell: first fragments, more fragments, in the queue, and pages not found yet. | | Last pages read | The last eight addresses with their status code, response time and a note: rendered, redirect or CAPTCHA. Click or Enter opens the page in the browser. | | Load by stage | Pages a second for each stage over the run. The lines start after the first seconds. | ## Views and Compact - **Overview** — the cards and charts above. - **Neuro-pipeline** — the same run drawn as a network of its stages. - **Terminal** — the same run drawn as a terminal, with the topics layer as tiles of dots. - **Compact** — in Overview only: each card keeps its numbers, the fragments shrink to a strip, and the run chart and **Load by stage** stay as small, low strips: the stages' pace and what held them, no more. The run fits beside other work, and the app remembers the choice. ## When the site asks for a CAPTCHA A site may answer with a bot check instead of its pages — in place, or by sending every page to a check page. The crawl then pauses, and a window opens with the site's check for you to pass. A bar over the app says **The crawl is paused: the site asks for a CAPTCHA.** Once the site lets you through, the crawl goes on by itself, with the check passed. | Button on the bar | What it does | | --- | --- | | Show the window | Brings the check's window back to the front. | | I passed it, continue | Tries the page again and resumes the crawl now. | | Skip checks in this crawl | Goes on without stopping at checks again; pages that answer with one are noted as CAPTCHA. Closing the check's window does the same. | [Settings › Crawling](https://www.getlooploop.com/gllmolfar/docs/settings-crawling) holds the switch, on by default: **At a CAPTCHA, pause and open it for me to pass**. Off, the crawl does not stop, and such pages are counted as met with a CAPTCHA. ## After the crawl When the crawl ends but the model is still reading fragments, the screen says **Stage 2 of 2 · finishing**: pages crawled and the time, fragments read out of the total, and about how long is left. **Ready now** lists the fixes that do not wait for the fragments. - **Embeddings off** — the app says: "Embeddings are off" Without them there are no topics, cannibalisation pairs or meaning search. Switch them on in the toolbar; the queue keeps what is waiting. - **No score yet** — the app says: "Scored when the crawl ends: the checks need every page." The score waits for the last page, so the screen does not guess it. - **Limit** — the app says: "This crawl stops at N URLs; M found so far." The crawl reached the URL limit set when it started. ## Related pages - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [Settings › Storage](https://www.getlooploop.com/gllmolfar/docs/storage.md): Settings › Storage: how much disk the saved crawls, the embedding model and downloaded updates take, per project, and the actions that free it. - [Reports](https://www.getlooploop.com/gllmolfar/docs/reports.md): The Reports tab: eleven reports in four directions, each answering one question about the open crawl, with a language filter and HTML or PDF export. - [Settings › Crawling](https://www.getlooploop.com/gllmolfar/docs/settings-crawling.md): Settings › Crawling: request speed, connections, the user agent, the timeout, robots.txt, a pause at a CAPTCHA, JavaScript rendering and service pages. --- # Overview > The first tab once a crawl has finished. It shows one score for the site, six counters, the fixes to make first, and what the site is about. Every number on it opens the pages behind it. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/overview - **Where:** Overview, the first section of the rail - **Mode:** Simple and Pro - **Built from:** The open crawl - **Changes:** When you open another crawl ## On the screen At the top, the score and six counters. Below them, two columns: on the left what to fix and how deep the site is; on the right the checks by importance, the pages, the topics and the earlier crawls. With Search Console connected, its figures join both columns. ![The Overview tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/overview.webp) Numbered parts of the screenshot: 1. **Site health.** A score from 0 to 100 with its band, and one sentence about the site with the number of things to fix. Under it: the site, the crawl's date and the number of HTML pages. 2. **Actions.** **All fixes (N)** opens the Fix list, **All checks** opens Issues, **Make a report** builds a PDF or HTML report of this crawl. 3. **Counters.** URLs crawled, indexable pages, errors, topics, cannibalization and pages rendered with JavaScript. Each opens the pages it counts. 4. **Fix first.** The first eight fixes, most important first: importance, the problem, how many URLs, and the share of pages it touches. A row opens the fix with its explanation. 5. **Checks and pages.** The checks found, by importance, then two bars: response codes and the indexability of HTML pages. Each part of a bar opens its pages. 6. **What the site is about.** The topics, each with its number of pages. A topic opens its pages; **Open the topic map** opens the map. 7. **Site depth.** HTML pages by clicks from the home page, and by folders in the address. A column opens its pages. ## Site health Each failed check takes away points by its importance: critical 5, high 3, medium 2, low 1. Insights take away nothing. A check takes away its full weight once it touches a third of the pages, and part of it below that. The score is the share of points left. | Score | Band | Sentence on screen | | --- | --- | --- | | 80–100 | good | {site} is in good shape. N things to look at. | | 55–79 | fair | {site} needs some work. N things to fix. | | 0–54 | poor | {site} has serious problems. N things to fix, most important first. | > **AI mode.** With the focus set to **AI** at the foot of the rail, the ring shows **AI readiness** instead, and the sentence says how well AI crawlers can get in and read the site. An **AI** button joins the actions. ## Counters | Counter | What it shows | Click opens | | --- | --- | --- | | URLs crawled | Every address fetched: pages, images, files, redirects. The line under it counts HTML pages. | All pages | | Indexable pages | HTML pages a search engine may index. The line under it counts the ones kept out of search. | The indexable pages | | Errors | Addresses that answered 4xx or 5xx, or did not answer. Red when above zero. | Those addresses | | Topics | Groups of pages by meaning. The line under it counts the pages embedded. A dash until topics are ready. | Topics | | Cannibalization | Pairs of pages that compete for the same meaning. Amber when above zero. | Cannibalization | | Rendered with JavaScript | Pages whose text needed a browser to appear. | Those pages | ## The two columns | Block | What it shows | | --- | --- | | Fix first | Up to eight fixes, the heaviest first: importance, the problem, the number of URLs, and a bar for the share of pages. **All fixes (N)** in its head when there are more. | | Lost in Google | Pages that lost Google clicks, worst first, with clicks before and after and the reason. Shown only with Search Console connected and only when something was lost. | | Site depth | Two charts of HTML pages: clicks from the home page, and folders in the address. A column opens its pages. | | Search Console | Clicks, impressions, CTR and position for the last 28 days against the 28 before, a line of daily clicks, and this week's new, lost, rising and falling queries. Not connected: a button to connect it. | | Checks found, by importance | How many checks failed at each level: critical, high, medium, low, insight. | | Pages | Response codes (2xx, 3xx, 4xx, 5xx, no response) and the indexability of HTML pages, each with the reason a page is kept out. | | What the site is about | The topics as chips, each with its number of pages. A chip opens the topic's pages; **Open the topic map** opens the map. | | Earlier crawls | How many times the site was crawled before and when last. **Open history** compares two crawls. | ## States and messages - **No crawl open** — the app says: "Enter a site address above and press Start." The start screen takes the tab's place, with the sites already crawled one click away. - **Crawl running** — the app says: "Scored when the crawl ends: the checks need every page." While a crawl runs, the tab shows the run instead. See [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen). - **No topics** — the app says: "Topics appear once pages are embedded. Turn on Embeddings if they are off." Topics and Cannibalization show a dash, and the topics block says why. ## Related pages - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. - [Reports](https://www.getlooploop.com/gllmolfar/docs/reports.md): The Reports tab: eleven reports in four directions, each answering one question about the open crawl, with a language filter and HTML or PDF export. - [Settings › Storage](https://www.getlooploop.com/gllmolfar/docs/storage.md): Settings › Storage: how much disk the saved crawls, the embedding model and downloaded updates take, per project, and the actions that free it. --- # Reports > Eleven reports drawn by the app from the open crawl. Each one answers one question, in four directions: technical SEO, content, AI visibility, and demand and progress. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/reports - **Where:** Reports, the second section of the rail - **Mode:** Simple and Pro - **Built from:** The open crawl, Search Console, history - **Export:** HTML, PDF ## On the screen ![The Reports tab of GLLmolfar with the Technical audit of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/reports.webp) Numbered parts of the screenshot: 1. **Report list.** The **Reports** section of the rail opens them in its panel, grouped by direction. The report open is highlighted. 2. **Language.** **All languages**, or one language version of the site with its number of URLs. 3. **Export.** **Save HTML** and **Save PDF** save the report as it is on screen. 4. **Beside your other projects.** The same figures for your other saved sites. With none saved, it says how to add one. 5. **Report head.** The direction, the report's name, the site, the crawl's date and the number of HTML pages, then the verdict in a few sentences. 6. **Key figures.** Four cards with the report's main numbers and how they moved since the last crawl. 7. **Details.** Charts and findings by level, each with the pages behind it. **Show these pages** opens them in the table. ## The eleven reports | Report | Direction | The question it answers | | --- | --- | --- | | Technical audit | Technical SEO | What is broken, where, and what to fix first | | Release check | Technical SEO | What changed since the last crawl, and did the release break anything | | Internal linking | Technical SEO | Which pages are deep, and which nobody links to | | Cannibalization | Content and optimisation | Which pages compete for the same meaning | | Stale content | Content and optimisation | Which pages are old, undated or unsourced, and which already lose clicks | | Structured data | Content and optimisation | Which templates carry which schema.org markup, and what is missing | | AI readiness | AI visibility | Can AI bots fetch and read the site: robots.txt, server answers, page structure | | Page citability | AI visibility | How a retriever might cut a page into passages, which read as quotable, and what falls past a 26-passage cap | | Question coverage | AI visibility | Which questions, from Search Console or your model, have a page on the site close in meaning | | Quick wins | Demand and progress | Which searches are a small change away from more clicks | | Monthly report | Demand and progress | What changed this month, what was done, and what comes next | Quick wins and Question coverage read Search Console and say so when it is not connected. Cannibalization and Question coverage need embeddings and say so when they are off. ## Language of a report On a site with several language versions, the language picker narrows a report to one of them: **Only the {l} pages: N of M URLs, with their Search Console pages.** Totals Google reports for the whole site stay as they are. Release check and Monthly report always cover all languages. ## States and messages - **No crawl** — the app says: "Open a saved crawl or start one: reports are drawn from it." Reports have nothing to read until a crawl is open. - **Loading** — the app says: "N s · usually about {t}" A report that loads data first (History, Search Console) shows the app's one wait bar: what it is reading, the seconds so far and, after the first time, how long it usually takes on this computer. - **Report beside others** — the app says: "Beside your other projects" Reports that compare show the same figures for your other saved sites. ## Related pages - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. --- # AI > What AI crawlers may fetch from the site and what they actually get, whether the pages read well for an assistant to quote, and which questions the site's passages answer. The bot check runs from this computer. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/ai - **Where:** AI › AI - **Mode:** Simple and Pro - **Built from:** The open crawl and a bot check of about 10 pages - **Changes:** When you press Check again or open another crawl ## On the screen At the top, the AI readiness score with its four parts. Below, two columns: on the left whether AI can get in and which questions the site answers; on the right whether AI can read and quote it, trust, llms.txt and Markdown for agents. ![The AI tab of GLLmolfar after a bot check of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/ai.webp) Numbered parts of the screenshot: 1. **AI readiness.** A score from 0 to 100 with its band. 2. **Headline and check.** One sentence about the site, when the bots were checked, and **Check again**. The note under it says how the check fetches. 3. **Four parts.** Access, Text bots get, Meaning and Trust, each with a bar. Meaning and Trust are marked **estimate**. 4. **Can AI get in.** Every AI bot with its purpose, whether robots.txt lets it into the home page, and what the sample pages answered. A bot opens its pages. 5. **Can AI read and quote it.** Lines with the number of pages each touches. Measured lines come first; style checks are marked **heuristic**. A line opens its pages. 6. **What the pages answered.** Pages fetched as that bot, by result. Bots that only exist as a robots.txt token say so. 7. **Who says it.** Authors, dates, unsourced numbers, the organization and its contacts across the site. ## AI readiness The app's own score, not a standard. Access (40%) and text bots get (30%) are measured: robots.txt, the sample pages fetched as each bot, text only after JavaScript and nosnippet. Meaning (15%) and trust (15%) are the app's estimate. The score never exceeds the lower of the two measured parts. A part not measured is left out and reads **not checked yet**. | Score | Headline on screen | | --- | --- | | 80–100 | AI crawlers can get in and read this site well. | | 55–79 | AI crawlers get in, but parts of the site are hard for them to read or quote. | | 0–54 | AI crawlers are blocked or see little of this site. | | Not checked | Check the site as AI bots to see how they get in. | > **Held down by a measured part.** When access or text bots get is below the rest, a line says so: “The score is held at N: it cannot be higher than Access, because bots that cannot get in or get the text cannot use the site, whatever its style.” ## Can AI get in **Check as AI bots** fetches about 10 pages, each as a browser and with the User-Agent of each AI bot robots.txt allows — from this computer, without JavaScript. A CDN that verifies a bot's IP address may answer the real bot differently. | Column | What it shows | | --- | --- | | Bot | GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, Claude-User, PerplexityBot, Perplexity-User, Google-Extended, Applebot-Extended, CCBot, Bytespider, meta-externalagent and Amazonbot, each with its owner. | | Purpose | **search** bots build the indexes AI search answers from; **user** bots fetch a page when someone asks; **training** bots only collect text for future models. | | Home page in robots.txt | **allowed** or **blocked**. A blocked training bot is grey and reads **your choice, does not affect AI search**. | | What the pages answered | Counts by result: like a browser (60%+ of its text), under 60% of the browser's text, CAPTCHA or bot check, error or refused, blocked by robots.txt. | A bot opens its pages: the address, what the page gave as a browser and as that bot (status and words), and the result. Under the table, **robots.txt, bot by folder** shows ✓ or ✕ for every bot in each folder; hovering gives the matching rule and its line. > **Training bots and Google.** Closing a training bot is a legitimate choice and does not remove the site from AI search. Google's AI Overviews and AI Mode use Googlebot, not Google-Extended. ## Can AI read and quote it | Line | Kind | | --- | --- | | AI sees under half the text (needs JavaScript) | Measured | | nosnippet or max-snippet: 0 | Measured | | noai or noimageai | heuristic | | Mostly menus and footers (under 40% content) | heuristic | | Sections that cannot be quoted alone | heuristic | | Walls of text (3,500+ characters without a heading) | heuristic | | Headings skip a level | heuristic | | Long text without lists, tables or headings | heuristic | | Has question headings (good) | A good sign, not a problem | Lines marked “heuristic” are style checks with no confirmed effect on AI answers. They do not move the score. ## The other blocks | Block | What it shows | | --- | --- | | Which questions the site answers | Paste the questions people ask, one per line, and press **Match questions**. The local embedding model matches each with every section of the site: **yes** from similarity 0.56, **weak** from 0.42, with the best section and its page. Thresholds set on one site: a guide, not a verdict. | | Questions people search | With Search Console connected: queries phrased as questions, with impressions, clicks and position, and a button to check them against the passages. | | Who says it, when, and with what sources | The share of long pages (600+ words, not the home page) that name an author, the share that show a date, and the share of pages with numbers without a source. Then the organization, sameAs profiles, About and contact pages, phones and emails; **pages state different values** when they disagree. Four lines open their pages: no author, no date, not updated for over 18 months, numbers and claims without a source. | | llms.txt | Whether `llms.txt` and `llms-full.txt` exist, their size, links and sections, links that lead to errors and links not in this crawl. When missing, a note on what the file is and that no major AI search engine has said it reads it. | | Markdown for agents | For each sample page: whether it answers `Accept: text/markdown` with Markdown, and whether a `.md` version exists, with the words in each. A warning when Markdown is served without `Vary: Accept`. | ## States and messages - **No crawl open** — the app says: "Enter a site address above and press Start." The tab needs a crawl. - **Not checked** — the app says: "AI bots have not been checked for this crawl." Access, llms.txt and Markdown say **Press Check as AI bots above.** - **Checking** — the app says: "Reading robots.txt…" A bar and the step: as a browser, as each bot, looking for llms.txt, asking for Markdown. - **robots.txt down** — the app says: "robots.txt did not answer (status N). Under RFC 9309 a crawler that cannot read robots.txt treats the whole site as closed, so AI bots may skip it entirely." Access counts as zero. - **Old crawl** — the app says: "Crawl again to read authors, dates and sources (crawls before 0.7.5 do not have them)." The trust block stays empty. > **Terms with a dotted underline.** Terms such as robots.txt, nosnippet and llms.txt carry a dotted underline. Hovering one shows a sentence on what it is and **More in the GetLoopLoop glossary**. [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general) switches it off. > **Locked screens.** A screen with a lock in the panel is not open on this account; pressing it says what opens it. ## Related pages - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [URL audit](https://www.getlooploop.com/gllmolfar/docs/url-audit.md): Site › URL audit: one page checked without a crawl — status, speed, JavaScript, page map, failed checks, accessibility, robots.txt, headers and screenshots. - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Agent · test](https://www.getlooploop.com/gllmolfar/docs/agent-test.md): The Agent · test screen: your own models plan tasks from the fix list, rewrite pages, write social posts and review risky pages; GLLmolfar measures the results. - [Reverse prompts](https://www.getlooploop.com/gllmolfar/docs/reverse-prompts.md): Site › Reverse prompts: GLLmolfar reads a page into blocks, your AI agent describes what each does, and every block gets a reusable prompt and an example. - [MCP tools for agents](https://www.getlooploop.com/gllmolfar/docs/mcp-tools.md): The MCP server inside GLLmolfar: how to connect an AI agent, and every tool it can call to crawl, read fixes, audit pages, search by meaning and save reports. --- # Agent · test > A test section where your own models do four jobs on the open crawl: a task plan, page drafts, social posts and a check for risky content. Every result says which model wrote it and what GLLmolfar measured or checked. Why a draft is better is written from the measurements, never by the model. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/agent-test - **Where:** AI › Agent · test - **Mode:** Simple and Pro - **Built from:** The open crawl and the models in Settings › AI models - **Changes:** Each time you run a job ## On the screen The heading carries a **test** mark and four sections: **Analysis and tasks**, **Content**, **Social posts** and **Risks**. Beside them, one chip per connected model role. Each section opens with a line on what it does, then its form and its result. ![The Agent · test screen of GLLmolfar on Analysis and tasks, with the getlooploop.com crawl open and no model connected](https://www.getlooploop.com/images/gllmolfar/docs/agent-test.webp) Numbered parts of the screenshot: 1. **Sections.** **Analysis and tasks**, **Content**, **Social posts**, **Risks**. The screen remembers the last one. 2. **What the section does.** One paragraph on the job and on what GLLmolfar checks afterwards. 3. **No model connected yet.** Shown until a model is set. **Open AI models** opens Settings › AI models. 4. **Goal (optional).** What the plan should aim at, in your words. 5. **Language.** English, Ukrainian, Spanish, French or Polish. 6. **What is sent.** How many fixes go to the model and which checks are left out, then **Build the plan**. ## Models The agent works on your own models: Ollama or LM Studio on this computer, or OpenRouter, OpenCode Zen, LiteLLM. They are connected and checked in [Settings › AI models](https://www.getlooploop.com/gllmolfar/docs/settings-ai-models). Each role shows as a chip in the heading — **Plans**, **Writes**, **Checks facts**, **Posts** — with the model and **On this computer or network** or **Cloud:** with its host. While a job runs: a bar, the step it is on, **step N of M**, the time, and the model's tokens as they arrive. **Stop** ends it. A finished job keeps one line: “Done in N s · model calls: N · tokens: N (in N, out N)”. Work is counted, never money. ## Analysis and tasks The model turns the crawl's fixes that have a consequence for search or visitors into tasks for the team, ranked by impact and effort. Length, format and header checks are not sent, and the line above the button counts them. | Part of a task | What it shows | | --- | --- | | Priority | As the model ranked it; a dash for a refuted task. | | Title and why | The task in the model's words. | | First step | When the model gives one. | | Effort and fixes | **effort** and the fixes it cites, by name. | | Verdict | **Matches the crawl**, **Partly matches** or **Refuted**, with the reason under the task. | > **Matched by GLLmolfar, not by the model.** Each task must cite fixes from this crawl, and the pages it states must lie between its largest fix and the sum of its fixes. A match says the task rests on real problems — not that the advice is right. **Fixes no task covers: N** lists the rest. **Copy Markdown** copies the tasks that were not refuted. ## Content **Improve a page** or **Blog article**. The model writes a draft with the skills you pick — Answer first, Rewrite for clarity, Add what is missing, FAQ, Title and description, Brand voice — and GLLmolfar measures it and sends the failures back, up to three drafts in all. Then a fact-checking call checks each claim against the source. | Block | What it shows | | --- | --- | | Form | The page to improve (with **Worth improving:** suggestions from thin pages and cannibalization pairs) or the page to build on, the topic and reader for an article, search queries to answer, notes for the writer, language, and **Write and measure**. | | Write → measure → rewrite | Each draft with its model, time and the checks it failed, or **All checks passed**. | | Measured by GLLmolfar | Now against Draft: words, passages against the app's cap of 26, where the best answer to the main query sits, fit to the main query, the closest page on the site, and passages that lean on the one before. Then each query with its impressions, before and after. | | Why this draft is better | Sentences built from the measurements only. When the best draft still fails a check: “Not ready yet … Do not publish it as is.” | | Draft | The best draft, with **Copy Markdown**. | | Fact check | Each claim marked ✓ supported, ✕ contradicted or ? not settled, with the source's words, and numbers that are not in the source. For a page, **The page as it is now** in a fold. | ## Social posts and Risks | Section | What it shows | | --- | --- | | Social posts | An article's address, then a row per network — X posts, X thread, LinkedIn post, LinkedIn article, Facebook, Threads, Bluesky, Telegram channel, Reddit — with a narrative (Serious, Hype, Story, Numbers first, Against the grain, A question, How-to) and, where it applies, how many posts or variants. Each post shows its length against the network's limit, numbers not in the article, and **Copy**. Nothing is posted: copy and publish yourself. | | Risks | **Check the site** re-reads every saved page of the open crawl for adult, pharma, gambling, scam, counterfeit or drug content, foreign script and pages that stand out in meaning; nothing is fetched. It lists **Pages to look at: N** with counts by category, the evidence on each page and a weight. **Ask the model about selected (N)** has your model judge each one as Harmful, Harmless mention or Unclear, and whether it looks hacked. | ## States and messages - **No model** — the app says: "No model connected yet" Connect one in Settings › AI models; **Open AI models** goes there. - **No crawl** — the app says: "Open a finished crawl first: the plan is built from its fix list." Analysis and tasks and Risks need a finished crawl. - **Working** — the app says: "Working…" The step on screen, such as “Ranking the fixes…” or “Checking every number against the crawl…”. - **No embeddings** — the app says: "not measured: the crawl has no embeddings" The closest page on the site cannot be measured. - **Fact check not run** — the app says: "The fact check did not run or failed: connect a model for “Fact checking” or any checked model, then run again. Numbers that are not in the source are flagged anyway." The draft is shown without the claim check. - **Nothing risky** — the app says: "Nothing found: no risky words, no foreign script, no page that stands out in meaning." The Risks list is empty. - **Stopped** — the app says: "The agent stopped: {detail}" The job ended with an error. ## Related pages - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. - [Reverse prompts](https://www.getlooploop.com/gllmolfar/docs/reverse-prompts.md): Site › Reverse prompts: GLLmolfar reads a page into blocks, your AI agent describes what each does, and every block gets a reusable prompt and an example. - [Settings › AI models](https://www.getlooploop.com/gllmolfar/docs/settings-ai-models.md): Settings › AI models: connect your own language models, on this computer, on your network or in a cloud, and choose which model does which analysis. - [Fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list.md): The Fix list tab: what to fix in the crawled site, numbered, most important first, each with why it matters, how to fix it and the URLs. - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization.md): The Cannibalization tab: groups of pages that say almost the same thing, the page to keep, the pages side by side and their text passage by passage. --- # Reverse prompts > Reverse prompting of a page and its blocks. GLLmolfar reads the page and cuts its copy into blocks; your AI agent, connected over MCP, writes what each block does; GLLmolfar checks the result and builds a prompt and an example for every block. It copies the shape of the text, never the text. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/reverse-prompts - **Where:** Site › Reverse prompts, or **Reverse prompts** in the URL audit - **Mode:** Simple and Pro - **Built from:** One page read by GLLmolfar, judged by your agent - **Changes:** When you read a page or your agent builds a run ## On the screen At the top, what the screen does, the address field and the state of the agent connection. Below, the pages read so far on the left and the selected one on the right. ![The Reverse prompts screen of GLLmolfar with the getlooploop.com address, before any page was read](https://www.getlooploop.com/images/gllmolfar/docs/reverse-prompts.webp) Numbered parts of the screenshot: 1. **What it does.** The page is read as HTML, as rendered and as its accessibility tree; your agent writes the judgements. 2. **Page address.** Filled with the open crawl's site. **Read the page** reads it and adds a run. 3. **Agent connection.** Whether your agent can do its part, with **How to connect**. 4. **Runs.** Every page read: blocks found, the address, the date, and **Built** or **Waiting for agent**. 5. **The run.** The selected run: its figures and, once built, the page profile and a prompt for every block. ## How a run goes - **Read the page.** GLLmolfar reads the address and cuts its copy into candidate blocks. The run appears in the list as **Waiting for agent**. - **Copy for the agent.** The run shows a line to paste into your AI agent (Claude Code, Codex or any MCP client connected to GLLmolfar): “Use GLLmolfar's reverse_prompts tools: get run {id} ({url}), fill in its analysis and build it.” - **Your agent writes the judgements** and sends them back through GLLmolfar's tools. GLLmolfar checks them. - **Built.** The prompts appear in the run, selected, as soon as the agent builds it. > **What GLLmolfar checks.** No six words in a row from the page and no brand name in anything the agent writes; a word range for every element; every candidate block placed once. Only a result that passes all five checks becomes prompts. ## A run | Part | What it shows | | --- | --- | | Figures while waiting | **candidate blocks**, **visible words**, **text only after JavaScript** and **text folded behind a click**, and **Read as:** with the layers read. | | Figures once built | **blocks with prompts**, **visible words** and **text only after JavaScript**. | | The page | **Main promise**, **Page and reader**, **Persuasion logic** and **Skeleton**: the page's blocks in order. | | Blocks (N) | Each block with its name, type and role, its **Prompt** with **Copy**, and an **Example from the prompt**. | | Actions | **Open the full report**, **Copy all prompts**, save as **.md**, **.json** or **.html**, and **Delete**. | ## The agent connection | Line on screen | What it means | | --- | --- | | The judgements are written by your AI agent, connected to GLLmolfar: AI agents need a signed-in account. | Sign in first. **How to connect** explains the rest. | | The judgements are written by your AI agent: switch the agent connection on in Settings and connect your agent. | Signed in, but the connection is off or not running. | | The agent connection is on: ask your agent to finish a run, or to reverse-prompt any page. | Ready. Your agent can also start a run itself. | ## States and messages - **No runs** — the app says: "nothing found" No page has been read yet. The right side says **Pick a page on the left.** - **Reading** — the app says: "Reading…" The button stays busy until the page is read. - **Bot check** — the app says: "The page shows a captcha or a bot check; GLLmolfar does not get past it." Nothing is read. - **Unreachable** — the app says: "The site did not answer or refused automated readers." Nothing is read. - **Not a page** — the app says: "Not a readable web page, or a page with no text (probably images)." Nothing to cut into blocks. - **Not public** — the app says: "Not a public web address." Nothing is read. - **Refused** — the app says: "Not analysed: a page in Russian, or a site in Russia's domain zone, hosted in Russia or with a Russian-language version. The publisher's decision." The page is not read. > **Locked screens.** A screen with a lock in the panel is not open on this account; pressing it says what opens it. ## Related pages - [URL audit](https://www.getlooploop.com/gllmolfar/docs/url-audit.md): Site › URL audit: one page checked without a crawl — status, speed, JavaScript, page map, failed checks, accessibility, robots.txt, headers and screenshots. - [Agent · test](https://www.getlooploop.com/gllmolfar/docs/agent-test.md): The Agent · test screen: your own models plan tasks from the fix list, rewrite pages, write social posts and review risky pages; GLLmolfar measures the results. - [MCP tools for agents](https://www.getlooploop.com/gllmolfar/docs/mcp-tools.md): The MCP server inside GLLmolfar: how to connect an AI agent, and every tool it can call to crawl, read fixes, audit pages, search by meaning and save reports. - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. --- # MCP tools for agents > GLLmolfar has a Model Context Protocol server inside it. An agent such as Claude Code, Codex or Hermes connects to it on this computer, starts crawls, reads the results and saves reports. Everything it does happens in the open window, where you can see it. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/mcp-tools - **Where:** Settings › General, and Help › Connect an AI agent… - **Reachable from:** This computer only, with a token - **Transport:** Streamable HTTP, or a command for clients that start a server - **Default:** Off until switched on ## Connect an agent - In [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general), switch on **Let AI agents use GLLmolfar on this computer (MCP)** and press **Save**. - Press **How to connect an agent…** beside it, or **Help › Connect an AI agent…**. - Copy the block for your agent with **Copy** and run or paste it where the dialog says. | Block in the dialog | Where it goes | What it holds | | --- | --- | --- | | Claude Code | Run once in a terminal: | A `claude mcp add --transport http gllmolfar …` command with the address and the token. | | Codex | Add to ~/.codex/config.toml: | A `[mcp_servers.gllmolfar]` entry that starts GLLmolfar's own bridge as a command. | | Hermes Agent | Add to ~/.hermes/config.yaml: | The same bridge, as `mcp_servers: gllmolfar:`. | | Any other MCP client | Streamable HTTP, or the command above for clients that start a server themselves. | The URL and the header `Authorization: Bearer `. | The address is `http://127.0.0.1:/mcp`. The port is normally `47863`; when another program already holds it, the app takes a free one, and the dialog shows the address in use. The bridge reads the port and the token from the data folder, so a Codex or Hermes setup needs no token pasted in. > **The token.** **Only programs on this computer can reach it, and only with the token. Anyone with the token can drive the app, so keep it to yourself; a new token cuts off every agent set up with the old one.** **New token** at the foot of the dialog makes one. The server also refuses requests that come from a web page in a browser. A lock means it is not open on this account; pressing it says what opens it. ## Crawls | Tool | What it returns or does | Kind | | --- | --- | --- | | `get_status` | What GLLmolfar does now: the open crawl, its progress, the account level and its URL limit, and whether Search Console data is there (`searchConsole`, with `say`: what to do). Agents are told to call it first. | Reads | | `list_crawls` | The saved crawls on this computer, newest first. Can filter by site. | Reads | | `open_crawl` | Opens a saved crawl in the window; the other tools then read it. | Changes the window | | `start_crawl` | Starts a crawl in the window and returns at once. | Fetches pages | | `wait_for_crawl` | Waits until the crawl, the embeddings and the topics are done, or the timeout. Returns the status. | Reads | | `pause_crawl` | Pauses the running crawl. The queue stays. | Changes the crawl | | `resume_crawl` | Continues a paused crawl from its queue. | Changes the crawl | | `stop_crawl` | Stops the running crawl. The pages crawled so far are kept. | Changes the crawl | | `get_history` | The saved crawls of one site with pages, errors, non-indexable pages and health, for trends. | Reads | | `compare_crawls` | Two crawls of one site: URLs found and lost and why, changes per URL, fixes that grew or went, robots.txt, AI bots, health and alerts. Without arguments: the open crawl and the one before. | Reads | ## Findings and fixes | Tool | What it returns or does | Kind | | --- | --- | --- | | `get_overview` | The open crawl in short: site, pages, health (0–100, GLLmolfar's score, not a Google score), topics, the number of fixes, and `searchConsole`. | Reads | | `get_fixes` | The [fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list), most important first. Each fix: key, importance, type, URL count, why, how, and `effect` (index, rank, ctr, users or none). `notCounted` names service pages a content check left out. Translation-only items come last, apart from health. | Reads | | `get_fix_urls` | The URLs of one fix key, also shown in the window. For broken links, redirects and similar keys, `linkedFrom` gives the pages that link to them, with anchor and position. The key `links` gives link opportunities. | Reads | | `explain_fix` | One fix key for a person who does not know the app: what it is, why it matters, how to correct it, how to verify it, with example URLs. | Reads | | `create_jira_task` | One fix as a full Jira ticket: fields, description in Jira markup and Markdown, a block for a coding agent, steps and Given/When/Then criteria, and attachments written as files. GLLmolfar sends nothing to Jira. | Writes files | | `find_risky_content` | Pages with risky content: words, foreign script and meaning unusual for the site. No model is used. The crawl must be stopped first. | Reads | | `get_ai_readiness` | The AI-readiness score (0–100) and its four parts. With `run_probe`, it fetches about ten pages as each AI bot and reports robots.txt per bot, llms.txt and Markdown for agents. | Fetches pages | ## Pages and meaning | Tool | What it returns or does | Kind | | --- | --- | --- | | `get_pages` | The open crawl's pages as a table: chosen columns, filters (fix key, topic, URL pattern, status, indexable, language), paged with `next_offset`. The column `fixes` lists each page's fix keys. | Reads | | `count_pages` | Page counts by one field, without rows: status, indexable, depth, topic, folder, template, language, page type, sitemap or fix. | Reads | | `audit_url` | One live page without a crawl: status, indexability, time, headers, redirects, JavaScript changes, robots.txt per AI bot, accessibility, headings, JSON-LD and the start of the text. Full text or HTML on request. | Fetches pages | | `semantic_search` | Pages and sections of the open crawl that answer a question, with scores. Needs embeddings. | Reads | | `get_topics` | The [topics](https://www.getlooploop.com/gllmolfar/docs/topics): label, terms, pages, cohesion, link coverage, cannibalization pairs, pillar page and example URLs. Needs embeddings. | Reads | | `get_cannibalization` | Groups of pages with almost the same meaning, with their pairs and similarity. The first URL of a group is the page to keep. Needs embeddings. | Reads | | `measure_text` | Measures a draft with embeddings on this computer, without a language model: its passages, the best passage per query, the nearest page of the crawl, and passages that lean on an earlier one. | Reads | | `get_search_console` | Search Console data for the open crawl's site: totals against the period before, top queries and pages, this week's movement, shared queries and pages that lost clicks. Search Console must first be connected in the app's [GSC](https://www.getlooploop.com/gllmolfar/docs/search-console) section. | Fetches data | ## Reports | Tool | What it returns or does | Kind | | --- | --- | --- | | `list_reports` | The reports of the [Reports](https://www.getlooploop.com/gllmolfar/docs/reports) section, each with an id, its direction, `ready` and, when not ready, `why`. | Reads | | `get_report` | Opens one report in the app and returns its verdict, figures and findings. Writes the same page as HTML or PDF to Documents\GLLmolfar\Reports, with a new name each time. Takes a `note` shown under the report's head, and the person's own projects to compare. | Writes files | | `save_report` | Saves the classic summary report of the open crawl to Documents\GLLmolfar\Reports and returns its path. Never replaces an older file. | Writes files | ## Planned crawls | Tool | What it returns or does | Kind | | --- | --- | --- | | `list_schedules` | The [planned crawls](https://www.getlooploop.com/gllmolfar/docs/schedule) with their next run, the queue, and the runs of the last 30 days. | Reads | | `save_schedule` | Adds a planned crawl or changes one by id: start as local time and repeat. Several URLs make one schedule each. | Changes schedules | | `run_schedule` | Runs one planned crawl now; if a crawl is running, it waits after it. | Fetches pages | | `delete_schedule` | Deletes one planned crawl by id. Its queued runs are skipped; saved crawls stay. Agents are told to ask first. | Deletes | ## Reverse prompts | Tool | What it returns or does | Kind | | --- | --- | --- | | `reverse_prompts_read` | Step 1 of 2. Reads a public page and splits its text into candidate blocks. Returns a run id, the brief, a draft analysis and the rules. | Fetches pages | | `reverse_prompts_build` | Step 2 of 2. Takes the finished analysis, checks it and builds a prompt and a measured example for each block, or names the lines to correct. | Changes the run | | `reverse_prompts_get` | One run by id: what it waits for, or the page profile and each block's prompt and example. | Reads | | `reverse_prompts_list` | The runs on this computer, newest first, with URL, status and run id. | Reads | ## Tools that use your own model These three use the model connected in [Settings › AI models](https://www.getlooploop.com/gllmolfar/docs/settings-ai-models). The server tells agents to call them only when the person asks. | Tool | What it returns or does | Kind | | --- | --- | --- | | `plan_fixes` | An action plan from the open crawl's fixes. GLLmolfar checks each task against its fixes and page counts. The result shows on the [Agent · test](https://www.getlooploop.com/gllmolfar/docs/agent-test) screen. | Uses a model | | `write_content` | Rewrites a page, or writes a blog post from it. Each draft is measured with embeddings, sent back for another round if it fails, and fact-checked against the page. Can take minutes. | Uses a model | | `write_social` | Posts about one page for each network asked for, each with its length, the network's limit and any numbers not in the page. | Uses a model | ## Recipes The server also offers named jobs as MCP prompts. Agents that support prompts show them as slash commands; **Help › Agent skills…** lists the same recipes in the app. | Prompt | Title | When | | --- | --- | --- | | `technical-audit` | Full technical audit → action plan | First contact with a site, or a quarterly audit. | | `ai-readiness` | AI-readiness audit of the whole site (GEO, AEO) | Can AI assistants read and cite the site? | | `page-agent-readiness` | Agent-readiness check of one page | Before launching a key landing, signup or pricing page. | | `citability` | Citability review and rewrite plan for one page | A page that should be cited for a set of questions but is not. | | `question-coverage` | Question fan-out coverage test | Planning content for a topic: which questions does the site not answer? | | `cannibalization` | Cannibalization cleanup | The cannibal check appears, or two pages split one query in Search Console. | | `internal-links` | Internal linking plan | Orphans, deep pages, or a new page that needs links. | | `llms-txt` | llms.txt draft | The site wants an llms.txt, or GLLmolfar flagged problems in the existing one. | | `refresh-brief` | Content refresh brief for one page | A page losing clicks, stale, or beaten by a competitor. | | `stale-content` | Stale, undated and unsourced content sweep | Quarterly E-E-A-T hygiene. | | `jira-backlog` | Jira backlog from a crawl | Handing technical work to developers. | | `release-check` | Release check: did this deploy break anything? | After a deploy, migration or redesign. | | `monthly-report` | Monthly client progress report | Agency reporting. | | `structured-data` | Structured data plan | Schema errors, or no schema on key templates. | | `fix-and-verify` | Fix it in the codebase, then verify | The agent is working inside the website's repository. | | `competitor-passages` | Competitor passage comparison | Why are they cited and we are not? | | `gsc-opportunities` | Search Console × crawl: quick wins | Quick wins with real demand behind them (Search Console connected in GLLmolfar). | | `reverse-prompts` | Writing prompts for a page type (reverse prompts) | Briefing writers or a model to produce more pages like the best one. | | `checked-audit` | Audit with live checks: do, do not, check first | Before you give a fix list to a person or a developer. | | `verify-fixes` | Verify the fixes after a deploy | After the developers say the fixes are live. | ## States and messages - **On** — the app says: "On: http://127.0.0.1:47863/mcp" The server is listening at that address. - **Starting** — the app says: "Starting…" Switched on; the server is not listening yet. - **Off** — the app says: "Off. Switch it on in Settings › General." No agent can connect. - **Not running** — the app says: "GLLmolfar is not running, or its agent connection (Settings › General) is off." From the bridge, to an agent that starts it as a command. - **Token changed** — the app says: "GLLmolfar answered 401: the token changed; restart the agent." From the bridge, after **New token**. ## Related pages - [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general.md): Settings › General: the interface language, term explanations, the interface mode, the default crawl size and mode, updates, statistics, the tray, MCP, sound. - [Settings › AI models](https://www.getlooploop.com/gllmolfar/docs/settings-ai-models.md): Settings › AI models: connect your own language models, on this computer, on your network or in a cloud, and choose which model does which analysis. - [Agent · test](https://www.getlooploop.com/gllmolfar/docs/agent-test.md): The Agent · test screen: your own models plan tasks from the fix list, rewrite pages, write social posts and review risky pages; GLLmolfar measures the results. - [Reports](https://www.getlooploop.com/gllmolfar/docs/reports.md): The Reports tab: eleven reports in four directions, each answering one question about the open crawl, with a language filter and HTML or PDF export. - [Fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list.md): The Fix list tab: what to fix in the crawled site, numbered, most important first, each with why it matters, how to fix it and the URLs. - [Schedule](https://www.getlooploop.com/gllmolfar/docs/schedule.md): The Schedule tab: a month calendar of planned and past crawls, the list of schedules, and the dialog that plans a crawl once or on repeat. --- # Projects > Every site crawled on this computer, as a card or a table row. Numbers compare the last crawl with the one before, and one line on each card says what to read first. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/projects - **Where:** Projects › Projects in the rail - **Mode:** Simple and Pro - **Built from:** The saved crawls on this computer - **Changes:** After each crawl, deletion or domain lookup ## On the screen At the top, four numbers for all sites together. Under them, a search box, filters, the order and the view. Then one card per site. ![The Projects tab of GLLmolfar with the card of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/projects.webp) Numbered parts of the screenshot: 1. **Summary.** **Sites**, **Average health**, **Need attention** and **Next planned scan**, for every site together. 2. **Toolbar.** **Find a site…**, the filters, **Sort** and the switch between **Cards** and **Table**. 3. **Card head.** The site, its number of crawls and the date of the last one, the health score with its change, the schedule, and **Main thing now:**. 4. **Four numbers.** **Pages**, **Errors**, **Not indexable** and **To fix** from the last crawl, each with its change since the crawl before. 5. **Scans and topics.** The last scans as columns: the height is the pages read, the line is health. Under them, the site's topics. 6. **Technical and Domain.** robots.txt, the sitemap and the user agent of the last crawl; the domain's age, the registrar and the days left before it expires. 7. **Actions.** **Open**, **Crawl again**, **History**, **Plan** and the bin, which deletes the project. ## The four numbers at the top | Tile | What it shows | Line under it | | --- | --- | --- | | Sites | How many sites have saved crawls. | {g} with Search Console, {s} on a schedule | | Average health | The average health score of each site's last crawl. | from each site's last crawl | | Need attention | Sites whose **Main thing now** is a problem that got worse. | The first such site and its line, or **nothing got worse since the crawl before** | | Next planned scan | The date of the next planned crawl of any site. | The site and the time, or **no scans planned** | ## Search, filters, order and view | Control | Options | | --- | --- | | Find a site… | Keeps the sites whose name contains the text. | | Filter | **All**, **Need attention**, **With Search Console**, **On a schedule**. | | Sort | **Needing attention first** (the default), **Health, lowest first**, **Last crawl**, **Name**. | | View | **Cards** or **Table**. The choice is kept while the app is open. | The table has one row per site: **Site**, **Health**, **Pages**, **Errors**, **Not indexable**, **To fix**, **Last crawl**, **Next scan** and **Delete**. The site's name opens its last crawl. ## One card | Part | What it shows or does | | --- | --- | | Icon | The site's own favicon; its first letter while it loads or when there is none. | | Health | The score of the last crawl, with ▲ or ▼ and the change since the crawl before. A dash until the crawl is opened once. | | Search Console | Shown when the site is connected to Search Console. | | Schedule | The next planned crawl, such as **Every day at 02:00**, which opens [Schedule](https://www.getlooploop.com/gllmolfar/docs/schedule); or **No schedule**. | | Main thing now: | One line about what to read first. See the table below. | | Pages, Errors, Not indexable, To fix | From the last crawl. Under each, the change since the crawl before, or **no change**. | | Scans | **the last N: a column is the pages read, the line is health**, with **up to N pages** for the tallest column. A column's colour is its health; a column opens that crawl. With one crawl: **Crawl again to see the trend.** | | Topics | The first four topics with their number of pages; **N more** shows the rest and **fewer** folds them. A topic opens the crawl and shows that topic's pages. | | Technical | robots.txt (the site's or custom, sitemaps, Disallow lines, Crawl-delay), the sitemap (**N URLs listed** or **not found**), the user agent the site was last crawled as, and CAPTCHA pages if there were any. | | Domain | The domain's age, the registrar and the days left. Opened: **Registered**, **Expires**, **Record changed**, **Registrar**, **Owner**, **Name servers**, **Server IP**, **Email (MX)**, **Protection** (DNSSEC and transfer lock) and where the record came from. ↻ looks it up again. | > **Domain lookups.** The app looks each domain up on its own, one at a time, when a card has no record or the record is a month old. **Registry record (RDAP or WHOIS) and DNS: only the domain name is sent.** ## Main thing now The app checks these in order and shows the first that applies. | Line on the card | When | | --- | --- | | The last crawl did not finish ({state}): its numbers are partial. | The last crawl was stopped, failed or was interrupted. | | N pages with errors, N more than the crawl before. | Errors grew. | | N pages not indexable, N more than the crawl before. | Non-indexable pages grew. | | Health fell by N since the crawl before. | The health score dropped. | | N pages answered with a bot check (CAPTCHA): the crawl did not see them. | Pages answered with a bot check. | | N pages answer with an error. | There are errors, but no more than before. | | Open this crawl once to get its health score. | The last crawl has no score yet. | | Nothing got worse. N fixes left on the list. | There are fixes on the list. | | Nothing got worse and nothing is on the fix list. | Nothing to fix. | More errors, more non-indexable pages and falling health are shown in red; those are what **Need attention** counts. ## Buttons | Button | What it does | | --- | --- | | Open | Opens the last crawl on the Overview. | | Crawl again | Starts a new crawl of the site's address. | | History | Opens the last crawl in [History](https://www.getlooploop.com/gllmolfar/docs/history). | | Plan | Opens **Plan a crawl** with the site's address filled in. See [Schedule](https://www.getlooploop.com/gllmolfar/docs/schedule). | | Delete project | Deletes every saved crawl of the site, its planned crawls and its domain data, after one confirmation. A Search Console link stays. | ## States and messages - **Empty** — the app says: "No crawls saved yet. Crawl a site and it will get its card here." Nothing has been crawled on this computer yet. - **No match** — the app says: "No site matches this filter." The search or the filter left no site. - **Looking up** — the app says: "Looking up…" The domain record is being read. - **Delete** — the app says: "Delete the project {site}?" Asks first, with the number of scans and of planned scans. After it: **Project {site} deleted: N scans.** - **Busy** — the app says: "Stop the running crawl first." A project, a topic or **Crawl again** cannot be used while a crawl runs. - **Topic gone** — the app says: "This topic is not in the crawl's analysis any more." The topic on the card was not found when the crawl opened. ## Related pages - [History](https://www.getlooploop.com/gllmolfar/docs/history.md): The History tab: every saved crawl of a site, a trend of one number across them, and a comparison of two crawls page by page and check by check. - [Schedule](https://www.getlooploop.com/gllmolfar/docs/schedule.md): The Schedule tab: a month calendar of planned and past crawls, the list of schedules, and the dialog that plans a crawl once or on repeat. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [Settings › Storage](https://www.getlooploop.com/gllmolfar/docs/storage.md): Settings › Storage: how much disk the saved crawls, the embedding model and downloaded updates take, per project, and the actions that free it. --- # Schedule > Crawls that run on their own, once or on repeat. The calendar shows the month's past runs and the planned ones; the list beside it holds every schedule with its next run. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/schedule - **Where:** Projects › Schedule in the rail - **Mode:** Simple and Pro - **Built from:** The schedules and the log of their runs - **Changes:** When a schedule is added, changed or runs ## On the screen ![The Schedule tab of GLLmolfar with the calendar of November 2026](https://www.getlooploop.com/images/gllmolfar/docs/schedule.webp) Numbered parts of the screenshot: 1. **Month.** ‹ and › move a month; **Today** comes back to this month. 2. **Legend.** **Planned**, **Running**, **Done**, **Failed** and **Missed**. 3. **Plan a crawl.** Opens the dialog that plans a crawl. 4. **Calendar.** Each day shows up to four runs, with the time and the site; more show as **+N**. A run opens its details; an empty spot of a day plans a crawl for that day. 5. **Schedules.** Every schedule with its repeat, its page count and its next run. With none yet, as here, it says how to plan one. 6. **If the computer is off at that time.** What to do with a run the computer was off for, and **Start with the system, in the tray**. ## Schedules The panel shows the number of schedules, **Running now: {site}** while a planned crawl runs, and **Waiting: N** with the sites in line. Each schedule shows: | Part | What it shows or does | | --- | --- | | Site | The site's favicon and name. | | Repeat | **Once: {date}, {time}**, **Every day at {time}**, **Every week: {days} at {time}** or **Every month on day {day} at {time}**, with **up to N pages**. | | Next run | **next: {date}**, or **no more runs**, or **switched off**. | | On | Switches the schedule on or off without deleting it. | | ▶ | **Run now**. While another crawl runs: **It will start when the crawl in progress finishes.** | | Edit | Opens **Change the planned crawl**. | | ✕ | Deletes the schedule after **Delete the schedule?** — **Crawls it already made stay in History.** | Under the list: **A run the computer was off for** — **Run it as soon as the computer is back**, **Skip it** or **Skip it and remind me** — and **Start with the system, in the tray**. These are the same settings as [Settings › Planned crawls](https://www.getlooploop.com/gllmolfar/docs/settings-planned). **Clear past runs from the calendar** removes the log of past runs. ## Plan a crawl | Field | What it takes | | --- | --- | | Sites | One address per line. **Several sites at the same time run one after another.** When a schedule is changed, the field is **Site**, one address. | | Repeat | **Once**, **Every day**, **Every week** or **Every month**. | | Date / Starting | The day of the run, or the first day of a repeat. A new plan starts tomorrow. | | Time | The time of the run; a new plan starts at 02:00. | | Days of the week | Shown for **Every week**: the days it runs. | | Max URLs | Pages per run; starts at **Default max URLs** from Settings. | | Mode | **Spider + sitemap**, **Spider only** or **Sitemap only**; starts at **Default mode** from Settings. | > **Settings at run time.** **The crawl uses the crawl settings in force when it starts (speed, user agent, rules, embeddings).** ## One run A run in the calendar opens **Planned crawl**: the site and its status, the date (**started by hand** for a run started with Run now), the repeat, and for a finished run **N pages · N errors · N non-indexable**. Its buttons are **Open the crawl**, **Run now**, **Edit**, **Delete** and **Close**. | Status | Meaning | | --- | --- | | Planned | Due later. | | Waiting | Due, and waiting for the crawl in progress to finish. | | Running | The crawl is running. | | Done | The crawl finished. | | Stopped | The crawl was stopped. | | Failed | The crawl did not start; the reason is shown. | | Missed | The computer or the app was off at that time. | | Skipped | The run was skipped. | ## States and messages - **Empty** — the app says: "Nothing planned yet. Press “Plan a crawl”, or click a day in the calendar." No schedule exists. - **Started** — the app says: "Planned crawl of {site} started, up to N pages" A planned crawl starts in the window like one started by hand. - **Login** — the app says: "Planned crawls run while GLLmolfar is open or in the tray. Start it with the system to keep them running after a restart." Shown while **Start with the system, in the tray** is off. - **Waiting** — the app says: "{site}: the planned crawl waits for the crawl in progress to finish" A run came due during another crawl. - **Missed** — the app says: "{site}: the run of {time} was missed — the computer or the app was off" What happens next depends on **A run the computer was off for**. - **Tray off** — the app says: "Closing the window quits GLLmolfar, so planned crawls cannot run. Turn on Settings › “The close button keeps GLLmolfar running in the tray”." See [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general). - **Locked** — the app says: "A lock means it is not open on this account." The tab then says what opens it, in place of the calendar. ## Related pages - [Projects](https://www.getlooploop.com/gllmolfar/docs/projects.md): The Projects tab: one card per site crawled on this computer, with its health, how the last crawl changed, its scans over time, topics and domain. - [History](https://www.getlooploop.com/gllmolfar/docs/history.md): The History tab: every saved crawl of a site, a trend of one number across them, and a comparison of two crawls page by page and check by check. - [Settings › Planned crawls](https://www.getlooploop.com/gllmolfar/docs/settings-planned.md): Settings › Planned crawls: what happens to a run the computer was off for, starting with the system in the tray, and notifications for planned crawls. - [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general.md): Settings › General: the interface language, term explanations, the interface mode, the default crawl size and mode, updates, statistics, the tray, MCP, sound. --- # History > The saved crawls of the open crawl's site. A trend shows one number across them; pick an older crawl as A and a newer one as B to see which pages appeared, went missing or changed, and which checks got better or worse. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/history - **Where:** Projects › History in the rail - **Mode:** Simple and Pro - **Built from:** The saved crawls of the open site - **Changes:** When you pick other crawls as A and B ## On the screen ![The History view of GLLmolfar with 5 saved crawls of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/history.webp) Numbered parts of the screenshot: 1. **Site and count.** The site and how many crawls of it are saved. 2. **Trend.** One number across the crawls, oldest to newest, and whether it got better or worse. 3. **Select and delete.** **Select all**, **Delete selected (N)** and **Delete all except the open one (N)**. 4. **Saved crawls.** One row per crawl. **Older (A)** and **Newer (B)** pick the two to compare. 5. **Pick A.** With only B picked, the comparison waits: **Pick an older crawl as A.** Once both are picked, it appears here. 6. **History in the rail.** Under **Projects**, with the number of saved crawls of the open site. ## Trend Pick the number: **Health**, **URLs**, **Errors** (the default), **Non-indexable**, **Problems to fix** or **Topics**. A dot is a crawl; a larger dot is the open one. The right side says **Better: {from} → {to}**, **Worse: {from} → {to}** or **No change**, from the first crawl to the last. ## Saved crawls | Column | What it shows | | --- | --- | | Checkbox | Selects the crawl for **Delete selected (N)**. **Tick scans to delete several at once**; when some are ticked, **Selected N of M** and **Clear selection**. | | Older (A) | The crawl to compare from. | | Newer (B) | The crawl to compare with. The open crawl is B when the tab opens. | | Started | When the crawl started; **open now** marks the open crawl. | | URLs | Addresses fetched. | | Embedded | Pages embedded for topics. | | Errors | Addresses that answered with an error; red above zero. | | Non-indexable | Pages kept out of search. | | No title | Pages without a title. | | State | **complete**, **stopped**, **failed**, **interrupted** or **running**. | | Bin | **Delete this scan**. | ## Comparing two crawls | Block | What it shows | | --- | --- | | Comparing {a} (A) with {b} (B). | The dates of the two crawls. | | Counts | **New URLs**, **Not in the newer crawl** with a count per reason, **Changed pages**, and one count per changed field. | | Changed pages | Per page: **Status**, **Title**, **Meta description**, **H1**, **Canonical**, **Indexability**, **Word count** and **Meaning** that changed, as before → after. Meaning says **The text now means something different (similarity N)**. | | New URLs | URLs crawled in B and not in A. | | Not in the newer crawl | URLs crawled in A and not in B, each with its reason, and **Check these URLs now (N)**. | | Checks between the two crawls | Each check with **A** and **B** (URLs it flags in each), **Fixed** and **New**. A number under Fixed or New exports those URLs. | | Export | **Export lost URLs (in A, not in B)** and **Export found URLs (new in B)**. | | Reason a URL is not in the newer crawl | Meaning | | --- | --- | | not reached (limit or still linked) | The newer crawl did not get to it. | | no longer linked, not checked | Nothing links to it any more. | | checked: still live | **Check these URLs now** found it answering. | | checked: removed (404/410) | It is gone. | | checked: redirects | It moved. | | checked: error or no answer | It failed when checked. | > **Missing is not deleted.** **Missing from a crawl does not mean deleted. Check these URLs to see which really return 404 or 410.** When the newer crawl stopped at its limit or was stopped, the note says many were simply not reached. **Checks between the two crawls** leaves out the checks that need topics, embeddings or the sitemap scan, since older crawls may not have them. Fixed: flagged in A, still crawled in B and no longer flagged. New: flagged in B but not in A. ## States and messages - **No site** — the app says: "Crawl a site to start its history." No crawl is open. - **One crawl** — the app says: "Two crawls of the same site are needed to compare." Only one crawl of the site is saved. - **No A** — the app says: "Pick an older crawl as A." No crawl older than B is picked. - **Short trend** — the app says: "This number is known for fewer than two crawls. Health, problems and topics are saved when a crawl is opened in 0.7 or later." Pick another number, or open older crawls once. - **Partial trend** — the app says: "N older crawls have no value for this number: open them once to fill it in." The line leaves those crawls out. - **Long lists** — the app says: "Long lists are cut to the first 500 rows; the counts above are complete." Use the export buttons for the full lists. - **No changes** — the app says: "No changes in the compared fields." Nothing in the compared fields changed between A and B. - **Delete** — the app says: "Delete this saved crawl?" Asks first: **Its pages and embeddings are removed from this computer. This cannot be undone.** Several at once: **Delete N saved scans?** - **Busy** — the app says: "Stop the running crawl first." The crawl that is running cannot be deleted. ## Related pages - [Projects](https://www.getlooploop.com/gllmolfar/docs/projects.md): The Projects tab: one card per site crawled on this computer, with its health, how the last crawl changed, its scans over time, topics and domain. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. - [Settings › Storage](https://www.getlooploop.com/gllmolfar/docs/storage.md): Settings › Storage: how much disk the saved crawls, the embedding model and downloaded updates take, per project, and the actions that free it. --- # Issues > Every check the crawl failed, in one list, most important first. Pick a check to see why it matters, how to fix it and exactly which URLs have it. The checks that passed are one click away. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/issues - **Where:** Checks › Issues - **Mode:** Pro - **Built from:** The open crawl - **Changes:** When you open another crawl ## On the screen On the left, the checks with findings. On the right, the one you picked: what it is, how many URLs it touches, why it matters, how to fix it, and where to look. ![The Issues tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/issues.webp) Numbered parts of the screenshot: 1. **Checks section.** The **Checks** section in the strip of sections; the panel beside it holds **Issues**, **Fix list** and **Batch fixes**. A red dot marks it while critical or high checks have findings. 2. **Type and importance.** **All**, **Issues**, **Opportunities** and **Potential issues**, each with its count. Under them, how many checks failed at each level: Critical, High, Medium, Low, Insight. 3. **Checks.** One row per check with findings: its name, its importance, the column set it belongs to and its type, the number of URLs and a bar for the share of pages. 4. **The check.** Importance, type and column set, the check's name, and three figures: **URLs affected**, the share **of N checked**, and the **weight in health score**. 5. **Actions.** **Open in table**, **Export URLs**, **Jira task** and **Export report**. 6. **Why and how.** **Why it matters** and **How to fix**, written for this check. 7. **Where to look.** The URLs with the problem, most linked first, with the columns that show it. ## The list of checks | Part | What it shows | | --- | --- | | Type buttons | **All**, **Issues**, **Opportunities**, **Potential issues**. Each shows how many checks of that type have findings; pressing one limits the list to it. | | Importance line | Critical, High, Medium, Low and Insight, each with the number of checks found at that level. | | Row | An icon for the type (**!** issue, **↗** opportunity, **?** potential issue), the check's name, a line with importance · column set · type, the number of URLs, and a bar for the share of the pages it was checked on. | | Order | By importance, then by the share of pages the check touches. | | Show passed checks (N) | At the foot of the list. Opens the checks that found nothing, each with a ✓. **Hide passed checks** folds them again. | | Importance | Weight in health score | | --- | --- | | Critical | 5 | | High | 3 | | Medium | 2 | | Low | 1 | | Insight | 0: shown, never weighed | > **The count on the tab.** The red number beside **Issues** in the panel counts the critical and high checks with findings, not every row in the list. While it is there, a red dot sits on **Checks** in the strip of sections, so it shows even when another section is open. > **In Simple mode.** In Simple mode the Checks section has no Issues tab, and its strip button no red dot. A link that leads here — **All checks** on the Overview, **Details** in Batch fixes — switches the app to Pro. ## The detail of a check | Block | What it shows | | --- | --- | | Head | Importance, type and the column set the check belongs to (Internal, Response codes, Images…), then the check's name. | | URLs affected | How many URLs fail the check. | | of N checked | Their share of the URLs the check applies to. | | weight in health score | The check's weight: 5, 3, 2, 1 or 0. | | Why it matters | What the problem costs. | | How to fix | What to change. | | Not counted | Shown when service pages were left out of the check: **Not counted: N service pages (privacy, terms, login…).** **Show them** lists them in the table. | | Where to look | The URLs with the problem, most linked first, with the columns that show it. A click on an address opens the page in the table. | Where the proof is a link — broken pages, redirects, insecure links — each row has **›**. Opening it lists the links themselves: **From** or **To**, **Anchor text**, **Where** and **Rel**. | Kind of check | Where to look shows | | --- | --- | | Most checks | A table of URLs, up to 200. **Showing 200 of N. Open in table or export for all.** below it when there are more. | | Duplicates (titles, descriptions, H1, H2, exact copies) | **N groups**, each with the value the pages share and its pages: Address, Indexability, Inlinks, Depth, Words. **Keep** marks the strongest page of each group: indexable, most linked, closest to the home page, longest. Up to 40 groups. | | Pages that mean the same | Groups of pages that compete, **N pages compete**, with the highest similarity and **Compare**, which opens the group in Cannibalization. | ## Actions | Button | What it does | | --- | --- | | Open in table | Opens the check's filter in [All pages](https://www.getlooploop.com/gllmolfar/docs/all-pages), on the column set it belongs to. | | Export URLs | Saves the affected URLs with the columns of the detail. For a duplicate check, one row per page with its group; for a check proved by links, one row per link. | | Jira task | Opens the check as a Jira ticket, its fields and files on the left. | | Export report | Saves a PDF about this one check: why, how, where the pages are, and every affected URL. | | Export (toolbar) | Every check with findings, one row per affected URL: Importance, Type, Check, Section, URLs affected, Address. | Arrow keys move through the list. A right-click on an address offers the same actions as in the table. ## States and messages - **No crawl open** — the app says: "Enter a site address above and press Start." The tab is empty until a crawl is open. - **Nothing found** — the app says: "Nothing to fix in this crawl." No check of the chosen type has findings. - **Nothing picked** — the app says: "Pick a check on the left to see why it matters, how to fix it and where to look." The detail pane before a check is chosen. > **Locked tab.** A tab with a lock is not open on this account; pressing it says what opens it. ## Related pages - [Fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list.md): The Fix list tab: what to fix in the crawled site, numbered, most important first, each with why it matters, how to fix it and the URLs. - [Batch fixes](https://www.getlooploop.com/gllmolfar/docs/batch-fixes.md): The Batch fixes tab: problems grouped by the one place that fixes many pages at once — the server, the shared layout, a template or a topic's texts. - [All pages (Internal)](https://www.getlooploop.com/gllmolfar/docs/all-pages.md): The All pages tab, called Internal in Pro mode: every crawled URL in one table, with filters, search, column sets, an overview panel and a page inspector. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. --- # Fix list > The checks that need work, written as tasks and numbered in the order to do them. Each task says how many URLs it touches, why it matters, how to fix it and where to look. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/fix-list - **Where:** Checks › Fix list - **Mode:** Simple and Pro - **Built from:** The open crawl - **Changes:** When you open another crawl ## On the screen On the left, the numbered tasks. On the right, the one you picked, laid out as in [Issues](https://www.getlooploop.com/gllmolfar/docs/issues): figures, why and how, and the URLs. ![The Fix list tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/fix-list.webp) Numbered parts of the screenshot: 1. **How many.** **N things to fix, most important first.** 2. **Tasks.** One numbered row per task: the task as a sentence, its importance and type, the number of URLs and a bar for the share of pages. 3. **The task.** Its number, importance, type and column set, the sentence, and **URLs affected**, **of N checked** and **weight in health score**. 4. **Actions.** **Open in table**, **Export URLs**, **Jira task** and **Export report**. 5. **Why and how.** **Why it matters** and **How to fix**. 6. **Where to look.** The URLs with the problem, most linked first. ## What is listed, and in what order - Every check with findings, except those rated **Insight**: they weigh nothing and are not tasks. - Ordered by importance first, then by the share of pages the check touches. - Internal links worth adding join the list as one low-importance opportunity once the topics are ready: **N internal links worth adding**. - An orphan in the sitemap that is already among the pages with no internal links is one problem, not two rows: that row is dropped when it names no other page. | Row part | What it shows | | --- | --- | | Number | The task's place in the order. The same number heads its detail. | | Sentence | The task with its count, such as **4 URLs answer with a 4xx error** or **3 pages get no internal links**. A check without its own sentence shows its name and **N URLs**. | | Second line | Importance and type: Issue, Opportunity or Potential issue. | | Number on the right | URLs affected. | | Bar | The share of the pages the check applies to. | > **The count on the tab.** The number beside **Fix list** in the Checks panel is the number of tasks. ## The detail of a task The same as a check's detail in [Issues](https://www.getlooploop.com/gllmolfar/docs/issues): the three figures, **Why it matters**, **How to fix**, a line about service pages left out when there are any, and **Where to look** — a table of URLs, groups of duplicates with **Keep** on the strongest page, or groups of pages that compete. **N internal links worth adding** has its own detail: **From**, **To**, **Suggested anchor** and **Similarity** for each link, and **Open Link opportunities** instead of the other buttons. | Button | What it does | | --- | --- | | Open in table | Opens the check's filter in [All pages](https://www.getlooploop.com/gllmolfar/docs/all-pages). | | Export URLs | Saves the affected URLs with the columns of the detail. | | Jira task | Opens the task as a Jira ticket, its fields and files on the left. | | Export report | Saves a PDF about this task: why, how, where the pages are, and every affected URL. | | Export (toolbar) | The whole list, one row per affected URL: Order, Task, Importance, URLs affected, How to fix, Address. | ## States and messages - **No crawl open** — the app says: "Enter a site address above and press Start." The tab is empty until a crawl is open. - **Nothing to fix** — the app says: "Nothing to fix in this crawl." No check outside Insight has findings. - **Crawl running** — the app says: "N things to fix, most important first." While a crawl runs the list follows it a few seconds behind; it is exact once the crawl ends. > **Locked tab.** A tab with a lock is not open on this account; pressing it says what opens it. ## Related pages - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. - [Batch fixes](https://www.getlooploop.com/gllmolfar/docs/batch-fixes.md): The Batch fixes tab: problems grouped by the one place that fixes many pages at once — the server, the shared layout, a template or a topic's texts. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [Link opportunities](https://www.getlooploop.com/gllmolfar/docs/link-opportunities.md): The Link opportunities tab: weakly linked pages, the pages closest to them in meaning that do not link to them yet, and an anchor for each link. --- # Batch fixes > The problems of the crawl, grouped by where one change fixes many pages: a server setting, the shared layout, a page template, or the texts of one topic. Each batch says where to change it, what to fix once, and which pages it covers. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/batch-fixes - **Where:** Checks › Batch fixes - **Mode:** Simple and Pro - **Built from:** The checks of the open crawl and its topics - **Changes:** When you open another crawl ## On the screen On the left, the batches, the one that fixes the most first. On the right, the batch you picked: where to change it, what to fix, and the pages. ![The Batch fixes tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/batch-fixes.webp) Numbered parts of the screenshot: 1. **How many.** **N batches: one change fixes many pages. Together they cover M problems on pages.** 2. **Batches.** Pages in the batch, its name, its cause and number of problems, the problems on pages it fixes, and a square coloured by cause. 3. **The batch.** Its cause, its topic when it has one, its name, and the figures: **pages**, **problems in common**, **fixed by one change**. 4. **Actions.** **Open in table**, **Export URLs** and **Export report**. 5. **Where and why.** **Where to change it** and **Why as a batch**. 6. **What to fix once.** Each problem of the batch with its importance, how many of the pages have it, how to fix it, and **Details**. 7. **Pages in this batch.** Each page with its title and a mark per problem. ## Kinds of batches | Name on screen | Cause | When it appears | Where to change it | | --- | --- | --- | --- | | Whole site: server and CDN settings | server | A server problem on most pages, in more than one folder. | Server, CDN or hosting settings (headers, redirects, caching, bot protection). Usually one config file or one dashboard. | | Whole site: shared layout (menu, footer) | shared layout | A layout problem on most pages, in more than one folder. | The shared layout every page uses: header, menu, footer, sidebars. | | Whole site: the base page template | template | A template problem on most pages, in more than one folder. | The base template or layout every page is built on (for example the Article block, the head with meta tags, the byline component). | | Whole site: linking and sitemap | linking | A linking or sitemap problem on most pages, in more than one folder. | Navigation, hub pages and the sitemap generator. | | Template {folder} | template | A folder of at least three pages where most of them share a problem. | The page template that renders that folder: one file in the CMS or the codebase sets these fields for every page in it. | | Top-level pages (/page) | template | The same, for pages at the root of the site. | The page template that renders them. | | Topic «{topic}»: {check} | content | One content problem on three or more pages of one topic. | The texts themselves: one brief for the editors covers every page in the batch. | | {folder}: {check} | content | The same, grouped by folder when the pages have no topic. | The texts themselves. | - Only checks on pages count, and not those rated Insight. - Batches are ordered by what they fix: the importance of each problem times the pages that have it, added up. - The number beside **Batch fixes** in the Checks panel is the number of batches. It appears once the crawl has ended. ## The detail of a batch | Block | What it shows | | --- | --- | | pages | Pages in the batch. | | problems in common | How many checks the batch groups. | | fixed by one change | Problems on pages the batch fixes: each check counted once per page that has it. | | pages in {folder} | For a template batch: every HTML page in that folder, including those without the problems. | | Where to change it | The place to make the change, by cause (see the table above). | | Why as a batch | **N problems on P pages come from one place. Fix it there once, then crawl again to check.** | | What to fix once | Each problem: importance, name, **N of P pages**, how to fix it, and **Details**, which opens that check in [Issues](https://www.getlooploop.com/gllmolfar/docs/issues). | | Pages in this batch | Address, Title, and a numbered column for each of the first six problems: **●** the page has it, **✓** it does not. A line under the table names the numbered problems. Up to 300 pages. | In a template batch, pages that do not share every problem are highlighted, with **N pages do not share every problem (highlighted): check whether they use a different template.** | Button | What it does | | --- | --- | | Open in table | Shows the batch's pages in [All pages](https://www.getlooploop.com/gllmolfar/docs/all-pages), with the batch's name as a filter you can clear. | | Export URLs | Saves Address, Title and a column per problem, **Yes** where the page has it. | | Export report | Saves a PDF about this batch. | | Export (toolbar) | Every batch, one row per page: Group, Batch, Cause, Pages in group, Address, Title, Issues on this page. | ## States and messages - **No crawl open** — the app says: "Enter a site address above and press Start." The tab is empty until a crawl is open. - **No batches** — the app says: "No batches: the problems found are on single pages. See Issues." No problem is shared widely enough to be fixed in one place. > **Content batches need topics.** Content problems are grouped by topic once the pages are embedded. Pages without a topic, and every page when topics are not ready, are grouped by folder instead. > **Locked tab.** A tab with a lock is not open on this account; pressing it says what opens it. ## Related pages - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. - [Fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list.md): The Fix list tab: what to fix in the crawled site, numbered, most important first, each with why it matters, how to fix it and the URLs. - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [All pages (Internal)](https://www.getlooploop.com/gllmolfar/docs/all-pages.md): The All pages tab, called Internal in Pro mode: every crawled URL in one table, with filters, search, column sets, an overview panel and a page inspector. --- # Topics > The site's pages grouped by what they say. One row per topic: how many pages it has, the words that mark it, how tight and how well linked it is, and the page that acts as its hub. Another crawled site can be compared topic by topic. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/topics - **Where:** Meaning › Topics - **Mode:** Simple and Pro - **Built from:** The open crawl's embeddings, made on this computer - **Changes:** Live during the crawl; final when it ends ## On the screen At the top, one line with the counts. Under it, the comparison with another site. Then a strip of every topic's share of pages, and the table of topics. ![The Topics tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/topics.webp) Numbered parts of the screenshot: 1. **Counts.** How many topics and pages, how many pages fit no topic, and whether this is the live or the final view. 2. **Compare with.** Pick another crawled site to compare its topics with these. Shown once the final view is ready. 3. **Share of pages.** One band per topic, as wide as its share of pages. A band opens the topic's pages in the table. 4. **Topics, Pages, Key terms.** The topic's name, its number of pages with a bar, and the words that mark it. 5. **Cohesion, Linked inside, Cannibal pairs.** How tight the topic is, how well its close pages link to each other, and how many of its pages compete. 6. **Central page.** The most linked page of the topic. It opens that page's details. 7. **Show pages, On the map.** **Show pages** opens the table of pages filtered to this topic. **On the map** opens the topic map with this topic in focus. > **Dotted words.** A term such as **embeddings** in the comparison line has a dotted underline: hovering it gives one sentence on what it means and a link to the GetLoopLoop glossary. **Settings › General** switches the explanations off. ## The table | Column | What it shows | | --- | --- | | Topics | The topic's name, made from its key terms, with its colour. The same colour marks the topic on the map and in every other tab. | | Pages | How many pages belong to the topic, with a bar against the largest topic. | | Key terms: | The words that set the topic apart, taken from its pages' titles and H1s. | | Cohesion | How close the topic's pages are to its centre in meaning, from 0 to 1. Higher is a tighter topic. | | Linked inside | Of the topic's pairs of pages that are close in meaning (similarity 0.80 or more), the share where one page links to the other. Red below 10%. **after crawl** until links are known. | | Cannibal pairs | Pairs of pages in this topic that say almost the same thing. Amber at 100 and above. See [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization). | | Central page: | The page of the topic with the most internal links pointing to it: the topic's hub. | | Show pages / On the map | Open the topic's pages in the table of [all pages](https://www.getlooploop.com/gllmolfar/docs/all-pages), or on the [topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map). | > **Live and final.** During a crawl the topics are drawn from the pages embedded so far: **Live view; it settles when the crawl ends.** When the crawl ends they are worked out once more over every page: **Final view.** Topics, link coverage and pairs can change between the two. The toolbar's **Export** saves the table: topic, key terms, pages, cohesion, linked inside, cannibal pairs and central page. ## Compare with another site The **Compare with** list holds the other sites crawled with embeddings. Both crawls must be embedded by the same model. Nothing is worked out until a site is picked. | Part | What it shows | | --- | --- | | The two sites in numbers | Health score, URLs crawled, indexable pages, errors, not indexable, problems to fix, topics and cannibal pairs, side by side. The better figure is green, the worse red. From each site's latest crawl with embeddings. | | Topics both sites cover · N | Topics matched by meaning, with their similarity and a bar of **Pages, yours : theirs**. **Closest pages** lists the page pairs nearest to each other, with their similarity. | | Only on {site} · N | Topics one site has and the other does not, with their pages and key terms. For the other site: subjects they cover and you do not, candidates for new pages. | ## States and messages - **No embeddings** — the app says: "Topics appear once pages are embedded. Turn on Embeddings if they are off." No page has been embedded yet. The **Embeddings** box in the toolbar turns the model on. - **Too few pages** — the app says: "Not enough embedded pages for topics yet." Some pages are embedded, but not enough to group. - **Waiting for the model** — the app says: "topics wait for the model: {n} first fragments left" Printed in the status bar while the model catches up with the crawl. - **Nothing to compare** — the app says: "No other crawled site with embeddings yet. Crawl one with embeddings on, then pick it here." The comparison needs a second site crawled with embeddings. - **Comparing** — the app says: "Comparing topics…" The comparison runs once per pick and is kept until the topics change. - **Locked** — the app says: "A tab with a lock is not open on this account; pressing it says what opens it." The tab keeps its place in the Meaning panel with a lock. ## Related pages - [Topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map.md): The Topic map tab: every embedded page as a dot, placed by meaning and coloured by topic, status, indexability or depth, with a card for each page. - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization.md): The Cannibalization tab: groups of pages that say almost the same thing, the page to keep, the pages side by side and their text passage by passage. - [Link opportunities](https://www.getlooploop.com/gllmolfar/docs/link-opportunities.md): The Link opportunities tab: weakly linked pages, the pages closest to them in meaning that do not link to them yet, and an anchor for each link. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. --- # Topic map > Every embedded page as a dot. Pages that say similar things sit close together, and each topic covers its own area. A dot opens a card with the page and the pages closest to it in meaning. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/topic-map - **Where:** Meaning › Topic map - **Mode:** Simple and Pro - **Built from:** The open crawl's embeddings and topics - **Changes:** Live during the crawl; redrawn more precisely when it ends ## On the screen A row of tools across the top, the map in the middle, and on the right the legend with a card for the page picked on the map. ![The Topic map tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/topic-map.webp) Numbered parts of the screenshot: 1. **Find a page on the map.** Type part of an address or a title. Pages that do not match fade; Enter zooms to the matches. 2. **Colour by.** Topic, Status, Indexability or Crawl depth. 3. **Topic layer, page titles.** **Topic layer** shades the area of each topic. **Page titles when zoomed in** labels the dots once you zoom in. 4. **Zoom and Fit.** − and + zoom, the figure is the zoom, **Fit** shows every page again. 5. **The map.** One dot per page. Close dots mean close meaning; a bigger dot has more internal links. Each topic carries a label with its number of pages. 6. **Legend.** The topics, largest first, with their pages. A topic puts the map's focus on it; **No topic** collects the pages that fit none. 7. **Page card.** Empty until a dot is clicked. Then: the page, its topic and the pages closest to it in meaning. ## Tools | Tool | What it does | | --- | --- | | Find a page on the map | Matches the address or the title. The count shows as **N matches · Enter to zoom**; Enter fits the view to them. | | Colour by | **Topic**: each topic's colour. **Status**: 200 OK, Redirect, Error or no response. **Indexability**: Indexable or Non-indexable. **Crawl depth**: clicks from start, 0 to 6+, and **Not linked (sitemap only)**. The legend changes to match. | | Topic layer | Soft coloured areas, one per topic. Drawn only when colouring by topic. | | Page titles when zoomed in | Labels the best linked pages with their titles once the zoom is high enough. | | −, +, Fit | Zoom out, zoom in, and show every page. Fit also clears a topic in focus. | Under the map: **Scroll to zoom · drag to move · click a dot for details · double-click to open it in the table**. With the map focused, + and − zoom, 0 fits, the arrow keys move and Esc clears the picked page and the focus. ## Dots, legend and card | Element | What it shows | | --- | --- | | A dot | One embedded page. Its size grows with the internal links pointing to it. Grey and smaller: the page fits no topic. | | Hover | The page's address, title, topic and inlinks. | | Topic label | The topic's name and its number of pages, placed in the middle of its pages. | | Legend, N topics | Every topic with its pages. Pressing one fades the others and fits the view to it; pressing it again clears the focus. | | Card | The full address, the title, the topic with the page's **fit**, then **Status · depth · inlinks · words**. **Open in table** opens the page's details; **Open in browser** opens the page itself. | | Similar pages | In the card: the six pages closest in meaning, each with its similarity from 0 to 1. A row moves the map to that page. | > **What the distance means.** Positions come from the pages' embeddings: two dots are close when the texts are close in meaning, whatever words they use. The axes have no units. During a crawl the layout is a quick one; when the crawl ends a more precise one is drawn, and the dots move to their new places. The toolbar's **Export** saves the pages by topic: topic, address, title, fit, whether the page is the topic's central page, words and inlinks. ## States and messages - **No embeddings** — the app says: "The map appears once pages are embedded." No page has been embedded yet. Turn on **Embeddings** in the toolbar if they are off. - **Live layout** — the app says: "N pages. Quick live layout; a more precise one is drawn when the crawl ends." The crawl is still running, or the final layout is not ready. - **Final layout** — the app says: "N pages. Close dots mean close meaning; size shows internal links. Click a dot to open the page." The layout worked out over every page once the crawl ended. - **Looking** — the app says: "Looking for pages with similar meaning…" The card is finding the closest pages. - **Locked** — the app says: "A tab with a lock is not open on this account; pressing it says what opens it." The tab keeps its place in the Meaning panel with a lock. ## Related pages - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization.md): The Cannibalization tab: groups of pages that say almost the same thing, the page to keep, the pages side by side and their text passage by passage. - [Structure](https://www.getlooploop.com/gllmolfar/docs/structure.md): The Structure tab: the crawled site drawn as a crawl tree, a directory tree or a force-directed diagram, coloured by topic, status, depth, speed or a check. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. --- # Cannibalization > Pages of one site that say almost the same thing, so they may compete for the same searches. They are joined into groups. A group opens with the page to keep, the three ways to resolve it, the pages side by side, and their text compared passage by passage. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/cannibalization - **Where:** Meaning › Cannibalization - **Mode:** Pro - **Built from:** The open crawl's embeddings and topics - **Changes:** Live during the crawl; final when it ends ## On the screen On the left, the groups, closest first. On the right, the group picked. **Visualisation** swaps both for charts of every group at once. The separate Cannibalization report is described under [Reports](https://www.getlooploop.com/gllmolfar/docs/reports). ![The Cannibalization tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/cannibalization.webp) Numbered parts of the screenshot: 1. **Groups and Visualisation.** How many groups and pages, and the switch between the list and the charts. 2. **The group.** Its importance, its type, its topic, and how many pages compete. 3. **Export URLs, Export report.** The group's pages and pairs as a table, or a report of this group. 4. **Figures.** Pages, the highest similarity in the group, and its number of pairs. 5. **Group list.** One row per group: its pages, the page to keep, the next two, the highest similarity and the topic's colour. 6. **Suggested main page.** The page to keep and why, next to three ways to resolve the group. 7. **Side by side.** Up to five pages of the group in columns, field by field. A field with the same text on every page is marked **same**. > **In meaning, and in Google.** This tab finds pages that say the same thing, from their text. [GSC › Cannibalization](https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization) finds pages Google actually showed for the same query, from Search Console. Pages found in both places compete in Google's results, not only in their text. ## What counts as a pair Two pages form a pair when their texts are close in meaning: similarity 0.94 or more, on a scale from 0 to 1. Pairs that share a page are joined into one group. A group's number is the highest similarity of its pairs: amber below 0.97, red at 0.97 and above, where the texts are near copies. - Both pages are in the same topic and in the same language. - Both pages are indexable, and neither has its canonical pointing at the other. - One page under two tracking tags, two pages of one paginated list, and a page with its own translation in another language folder are not pairs. When there are very many, the analysis keeps the strongest pairs. The list then ends with **Showing {n} of {total}. Export CSV for the full list.**, and the export notes that the analysis kept the {n} strongest of {total} similar pairs. ## A group, on the right | Block | What it shows | | --- | --- | | Heading | **High** and **Potential issue**, the topic, and **N pages compete for the same searches**. | | pages · highest similarity · pairs | The size of the group, its closest pair (red at 0.97 and above), and how many pairs it has. | | Suggested main page | The strongest page, marked **Keep**: indexable, most linked, closest to the home page, longest. **It has** the most inlinks, the fewest clicks from home, the most text, where it does. When no page is clearly stronger: **No page is clearly stronger: pick the one that best answers the search.** | | Merge and redirect | Move what is unique on the others into the main page, then 301 them to it. | | Canonical | When the others must stay for visitors, point their canonical at the main page. | | Make them different | When each should rank for its own search, give each its own title, H1 and passages that only it has. | | Side by side | Title, H1, Meta description, Words, Inlinks, Depth, Indexability, Canonical, In sitemap and Response, ms, for up to five pages. A text field equal on every page is shaded and tagged **same**. A larger group says how many more pages are in the list. | ## What they say, passage by passage Under the table, two pages of the group are read and cut into passages, and each passage is matched with the closest passage on the other page. With more than one pair, chips above it pick the pair, each with its similarity. | Element | What it shows | | --- | --- | | of A repeated on B, of B repeated on A | The share of each page's text that the other page also says. | | page vectors alike | The similarity of the two pages as a whole. | | Words both use, Only on A, Only on B | The words the pages share, and the words only one of them has. | | says the same | Red: the passage says what a passage on the other page says. It names that passage's number and the similarity. | | close | Amber: a passage on the other page is close to it. | | only here | Green: no passage on the other page says this. This is what makes a page worth keeping. | Pointing at a passage lights the passage it matches. The line under the columns says where the text came from: **Text from the saved HTML.**, or that the page was requested once more to read its text, and **Compared with the built-in model on this computer.** ## Visualisation | Block | What it shows | | --- | --- | | Figures | Groups; Pages that compete, with their share of embedded pages; Pairs; Near-copy pairs. | | Cannibalization map | One circle per group, in rows by topic. Its pages sit on the rim; a line joins two pages that say almost the same, red for near copies (0.97 and above). Click a group to pick it, click again to open it in the list. | | How close the pairs are | Pairs by the similarity of their text. Red bars are near copies. | | Group size | Groups by how many pages compete in them. | | Inside a group | The picked group as a grid: each cell is the similarity of two of its pages; an empty cell is a pair below the threshold. **Open in the list** opens the group. | | Cannibalization by topic | Pages that compete with another page, per topic, and their share of the topic. | The toolbar's **Export** saves every group with its pages and pairs. ## States and messages - **No embeddings** — the app says: "Pairs appear once pages are embedded." No topics have been worked out yet. Turn on **Embeddings** in the toolbar if they are off. - **None found** — the app says: "No pages that mean the same. Good." No two pages pass the rule above. - **Embeddings off** — the app says: "Embeddings are off. Turn them on to compare by meaning." Shown in place of the passages: the comparison passage by passage needs the model. - **Locked** — the app says: "A tab with a lock is not open on this account; pressing it says what opens it." The tab keeps its place in the Meaning panel with a lock. ## Related pages - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Reports](https://www.getlooploop.com/gllmolfar/docs/reports.md): The Reports tab: eleven reports in four directions, each answering one question about the open crawl, with a language filter and HTML or PDF export. - [Topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map.md): The Topic map tab: every embedded page as a dot, placed by meaning and coloured by topic, status, indexability or depth, with a card for each page. - [Link opportunities](https://www.getlooploop.com/gllmolfar/docs/link-opportunities.md): The Link opportunities tab: weakly linked pages, the pages closest to them in meaning that do not link to them yet, and an anchor for each link. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. --- # Link opportunities > Internal links worth adding. Each row is one link: the page to put it on, the page it points to, an anchor, and how close the two pages are in meaning. The pages that need links most come first. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/link-opportunities - **Where:** Meaning › Link opportunities - **Mode:** Pro - **Built from:** The open crawl's embeddings and internal links - **Changes:** Once, when the crawl ends ## On the screen One line on what the tab shows, then one table. ![The Link opportunities tab of GLLmolfar after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/link-opportunities.webp) Numbered parts of the screenshot: 1. **What it shows.** Pages close in meaning that do not link to each other. The target has few internal links. 2. **Add a link on, Link to.** The page to edit, and the page the new link points to. Either opens that page's details. 3. **Suggested anchor.** Text for the link, taken from the target's H1, or its title when it has none. 4. **Similarity, Target inlinks, Topic.** How close the two pages are, how many internal links the target has now, and its topic. 5. **Order.** Targets with the fewest inlinks first; for the same count, the closest pages first. ## Columns | Column | What it shows | | --- | --- | | Add a link on | The page to put the link on. It is close in meaning to the target and does not link to it yet. | | Link to | The weakly linked page that needs the link. | | Suggested anchor | The target's H1, or its title when it has no H1, cleaned up for use as link text. | | Similarity | How close the two pages are in meaning, from 0 to 1. Only pairs at 0.80 or more are suggested. | | Target inlinks | Internal links pointing to the target now. | | Topic | The target's topic, with its colour. | ## Which pages get suggestions - Both pages are indexable and in the same topic. - The target has fewer than 5 internal links pointing to it. A page with more gets no suggestions. - Each target gets at most three pages to link from: the closest ones, with pages that already carry more link weight preferred. - Suggestions are worked out once, when the crawl ends. During a crawl the tab waits. When there are very many, the analysis keeps the strongest. The table then ends with **Showing {n} of {total}. Export CSV for the full list.** The toolbar's **Export** saves the suggestions the analysis kept, with a note that it kept the {n} strongest of {total}. ## States and messages - **Waiting** — the app says: "Link opportunities appear when the crawl ends and pages are embedded." The crawl is still running, or the pages are not embedded yet. Turn on **Embeddings** in the toolbar if they are off. - **None found** — the app says: "No missing links found." No weakly linked page has a close page that does not link to it yet. - **Locked** — the app says: "A tab with a lock is not open on this account; pressing it says what opens it." The tab keeps its place in the Meaning panel with a lock. ## Related pages - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization.md): The Cannibalization tab: groups of pages that say almost the same thing, the page to keep, the pages side by side and their text passage by passage. - [Fix list](https://www.getlooploop.com/gllmolfar/docs/fix-list.md): The Fix list tab: what to fix in the crawled site, numbered, most important first, each with why it matters, how to fix it and the URLs. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. --- # Semantic search > Search the crawled site by meaning. The query is embedded on this computer and compared with every embedded page, so a page can match without sharing a word with the query. Each result shows the parts of the page that match best. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/semantic-search - **Where:** Meaning › Semantic search - **Mode:** Pro - **Built from:** The open crawl's page and section embeddings - **Changes:** On each search ## On the screen A search box across the top, a line on how the search works, then the results. ![The Semantic search screen of GLLmolfar: the query how to measure AI visibility and the getlooploop.com pages that match it, each with its score and the sections that matched](https://www.getlooploop.com/images/gllmolfar/docs/semantic-search.webp) Numbered parts of the screenshot: 1. **Find pages by meaning.** The query: a question or a phrase, in any wording. **×** clears it. 2. **Search.** Runs the search. Enter does the same. 3. **How it works.** **The query is embedded on this computer and compared with every embedded page.** 4. **What was compared.** How many pages the query was compared with, and how many had their sections searched: a result shows the part of the page that matched. 5. **The page's score.** The similarity of the page as a whole to the query, and whether the page as a whole or **a section matches**. 6. **Matching sections.** The sections of that page closest to the query, each with its own score and the start of its text. | Part | What it shows | | --- | --- | | Find pages by meaning | The query: a question or a phrase, in any wording. | | Search | Runs the search. Enter does the same. | | How it works | The query is embedded on this computer and compared with every embedded page. | | Results | How many pages were compared, then the closest pages, best first, or a message saying why there are none. | ## A result Up to 25 pages, closest first. Above them: **Compared with N pages.**, and either **Sections searched on N pages: a result shows the part of the page that matched.** or, for a crawl made before 0.7.2, **This crawl has no section vectors (crawled before 0.7.2): crawl again to search by section.** | Element | What it shows | | --- | --- | | Number | The result's place in the list. | | Title and address | The page's title, or its address when it has none, and the full address with the colour of its topic. | | Score | How close the page is to the query, from 0 to 1, with three decimals and a bar. The bar is empty at 0.20 and full at 0.80. | | a section matches / the page as a whole | Which part gave the score: one section of the page scored higher than the whole page, or the page as a whole did. | | Sections | Up to three sections of the page, best first: each with its score, its heading (**Start of the page** for text before the first heading) and the start of its text. | A page scores as its best part: the whole page, or its best section. A result opens the page's details. The toolbar's **Export** saves the results on screen: rank, query, address, title, topic, similarity, best section and its text. > **Reading the score.** The score is the similarity between the query and the page in the embedding model on this computer. It ranks the pages for this query; it is not a percentage. ## States and messages - **No embeddings** — the app says: "No embedded pages yet." The open crawl has no embedded pages. Turn on **Embeddings** in the toolbar and let the model work through the pages. - **Locked** — the app says: "Not included in your level. Open the tab or menu item to see what includes it." Printed in place of the results. A tab with a lock is not open on this account; pressing it says what opens it. ## Related pages - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map.md): The Topic map tab: every embedded page as a dot, placed by meaning and coloured by topic, status, indexability or depth, with a card for each page. - [Reverse prompts](https://www.getlooploop.com/gllmolfar/docs/reverse-prompts.md): Site › Reverse prompts: GLLmolfar reads a page into blocks, your AI agent describes what each does, and every block gets a reusable prompt and an example. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. --- # All pages (Internal) > Every URL of the crawl in one table. Filter it by a check, search it, sort it, and pick a row to inspect that page below. In Pro mode the same table opens other column sets, compares with an earlier crawl and carries an overview panel. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/all-pages - **Where:** Site › All pages (Internal in Pro) - **Mode:** Simple and Pro - **Built from:** The open crawl - **Changes:** Live during a crawl ## On the screen From the top: the column sets, the filter bar, the table with the overview panel beside it when it is open, and the inspector for the picked page. With dozens of columns, this table scrolls sideways like a spreadsheet, where most tables in the app wrap or stack their rows to fit the window. ![The Internal tab of GLLmolfar in Pro mode after a crawl of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/all-pages.webp) Numbered parts of the screenshot: 1. **Column sets.** **Internal**, **Response codes**, **URL**, **Page titles**, **Meta description**, **H1**, and **More columns…** for the rest. Pro mode. 2. **Filter and compare.** **Filter**: all URLs or one check, each with its count. **Compare with**: an earlier crawl of the site. 3. **Search and tools.** Search by address or title, in all columns or one, with `.*` for a regular expression; then **Tree**, **Columns** and **Export**. 4. **The table.** **Showing N of M** above it. One row per URL; a header sorts, drags and resizes its column. 5. **Overview panel.** Opens and closes the panel of **Overview**, **Site structure** and **Response times** beside the table. Closed in this picture. 6. **Inspector.** The picked page: its address and state, and its tabs, from **URL details** to **About the page**. ## Simple and Pro | Part | Simple: All pages | Pro: Internal | | --- | --- | --- | | Column sets | Internal only | All of them | | Filter, search, Export, Showing N of M | Yes | Yes | | Compare with | — | When the site has an earlier crawl | | Search in one column, regular expression | — | Yes | | Tree, Columns, Overview panel | — | Yes | | Inspector | Yes | Yes | ## The Internal columns | Column | What it shows | | --- | --- | | Address | The URL. | | Status | The HTTP response code. | | Indexability | Indexable, or why the page is kept out of search. | | Topic | The page's topic, with its colour, once pages are embedded. | | Title, Title len. | The page title and its length. | | Meta len. | The length of the meta description. | | H1 | The first H1 heading. | | Words | Words on the page. | | JS | Whether the page was rendered with JavaScript. | | Depth | Clicks from the start page. | | Inlinks | Internal links to the page. | | Page type, Service page | What kind of page it is, and whether it was taken for a service page. | | Response, ms; Size, KB | Time to answer and size of the response. | | In sitemap | Whether the sitemap lists it. | | Embedded | Whether the page has an embedding. | **Columns** opens **Columns · {set}**: tick the columns to show and drag them into order. **Reset to default** brings back the set's own columns. With a crawl to compare, **Change since compared crawl** joins the table. The other sets under **More columns…**: H2, Content, Canonicals, Directives, Hreflang, JavaScript, Links, Sitemaps, Structured data, Images, External, Accessibility, AI, Service pages, Security, Headers & cookies, Pagination, and Custom when extraction rules are set. ## Filter, search and compare | Control | What it does | | --- | --- | | Filter | **All URLs (N)**, then each check with its count and share. Other column sets have their own filters. | | Compare with | **None.** or an earlier crawl, by date and URL count. Picking one adds the Change column and three filters: **New since the compared crawl**, **Changed since the compared crawl**, **Unchanged**. | | Search | **Address or title contains…**. **All columns** or one column; `.*` reads the search as a regular expression. | | Tree | Groups the rows by host and folder. A folder row shows **N URLs** and **N with errors**; a click opens or folds it. | | Showing N of M | Rows shown out of the URLs in the set; **N selected** when several rows are selected. | | Export | Saves the rows on screen with the visible columns. | A filter set from elsewhere — a topic, a depth column, a batch — shows as a chip beside the filter, with **✕** to clear it. ## Working with rows - Click picks a row; Shift and Ctrl add to the selection; Ctrl+A selects all. - Double-click or Ctrl+Enter opens the page in the browser. - Ctrl+C copies the URLs; Ctrl+Shift+C copies the rows with headers, to paste into a spreadsheet. - Del removes the selected URLs from the saved crawl, after asking. The site itself is not touched. - Right-click: Open in browser, Copy URL, Copy rows, Show inlinks, Show outlinks, View source, SERP snippet, Quick audit, Mark as service page…, Re-crawl, Remove from crawl, index checks in Google and Bing, Wayback Machine, Rich Results Test, PageSpeed Insights, W3C HTML validator, Open robots.txt, Export selected. ## Overview panel | Tab | What it shows | | --- | --- | | Overview | Every column set with its count, and under it each filter with its count and share. A filter opens it in the table. | | Site structure | **Top folders**: Folder, URL, Indexable pages, Errors. Then **Clicks from the home page** and **URL nesting (folders in the address)** as columns. A folder or a column filters the table. | | Response times | **Average N ms, median N ms.** and the URLs by response time, from 0–0.2 s to 5 s and more. A band filters the table. | ## Inspector Its head shows the picked page: favicon, address, status, indexability, service page and topic, with **↑** and **↓** for the previous and next row, **Open ↗**, **Bottom** or **Right** for where it sits, and **—** to fold it. | Tab | What it shows | | --- | --- | | URL details | Response, indexing and content of the page: status, redirects, canonical, robots, depth and links, hreflang, title and description with their length, headings, words, topic, page type, schema. | | Inlinks, Outlinks | Links into and out of the page: From or To, anchor text, Where, Rel, Status. | | Images | The page's images and their alt text: **N images, M without alt.** | | SERP snippet | The result as a search engine would show it, with the title and description editable and measured in pixels. | | Source | **Raw HTML**, **After JavaScript**, and the lines JavaScript added, with a search and **Copy**. | | HTTP headers | The response headers, grouped, and the security headers present or missing. | | Structured data | The schema types and each JSON-LD block. | | Rendered page | A screenshot of the page, **Desktop** or **Phone**. | | Similar pages | The pages closest in meaning, with their similarity; 1.000 means the same meaning. | | About the page | What the page is about, passage by passage, read by the model on this computer. | ## States and messages - **No crawl open** — the app says: "Enter a site address above and press Start." The table is empty until a crawl is open or running. - **No row picked** — the app says: "Click a row to see its details and similar pages." The inspector's head says **Pick a row to inspect its page**. - **Several rows** — the app says: "Several rows selected. Select one to see its details." The inspector shows one page at a time. - **Comparing** — the app says: "The Change column and the New / Changed filters show what is different from that crawl." Shown once when a crawl is picked in **Compare with**. > **Locked tab.** A tab with a lock is not open on this account; pressing it says what opens it. ## Related pages - [Structure](https://www.getlooploop.com/gllmolfar/docs/structure.md): The Structure tab: the crawled site drawn as a crawl tree, a directory tree or a force-directed diagram, coloured by topic, status, depth, speed or a check. - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. - [URL audit](https://www.getlooploop.com/gllmolfar/docs/url-audit.md): Site › URL audit: one page checked without a crawl — status, speed, JavaScript, page map, failed checks, accessibility, robots.txt, headers and screenshots. - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. --- # Structure > The site as a diagram. Each page is a dot on the path a visitor clicks to reach it, or on the folders in its address. Colour the dots by topic, status, indexability, depth, response time or one check, and click a page to see its numbers and its checks. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/structure - **Where:** Site › Structure - **Mode:** Simple and Pro - **Built from:** The pages and internal links of the open crawl - **Changes:** When a crawl ends or another opens ## On the screen A row of controls on top, the diagram in the middle, and on the right the legend and the card of the page you clicked. ![The Structure tab of GLLmolfar with the radial directory tree of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/structure.webp) Numbered parts of the screenshot: 1. **Diagram and layout.** **Crawl tree**, **Directory tree** or **Force-directed**; **Radial** or **Left to right** for the two trees. Here: the directory tree, radial. 2. **Colour, size, levels.** **Colour by**, **Size** and **Levels**. **All internal links** joins them in the crawl tree; the directory tree does not have it. 3. **Find.** **Find a page on the map**: matching pages get a ring, the rest fade; **N found** beside it. 4. **Zoom and PNG.** **−** and **+** with the zoom level, **Fit** to show the whole diagram, **PNG** to save it as an image. 5. **Counts and the tree.** Pages, folders and levels of the directory tree on screen. Below them, the folders in the address around the site: a folder without a page of its own is a grey square; an address with a query string hangs from its folder. 6. **Legend.** What each colour means. In topic colours, the first twelve topics; a grey square for groups or folders. 7. **Page card.** The clicked page: its numbers, the checks it fails, and what to do next. Empty until a page is clicked. ## Three diagrams | Diagram | What a branch is | Layout | | --- | --- | --- | | Crawl tree | The shortest click path from the start page. Pages reached without a link from a crawled page hang from the group **Not linked (sitemap or list only)**. | Radial or Left to right | | Directory tree | The folders in the address. A folder without a page of its own is a grey square; an address with a query string hangs from its folder. | Radial or Left to right | | Force-directed | The crawl tree, let loose: each page pulls toward the page above it, and all dots push each other apart. A dot can be dragged. | — | A red ring marks a page that answered 4xx, 5xx or nothing. A number beside a dot, **+N**, counts the pages folded under it by **Levels**. ## Controls | Control | Choices | What it changes | | --- | --- | --- | | Colour by | Topic, Status, Indexability, Crawl depth, Response time, A check | The colour of each dot. **A check** adds a list to pick one, each with its count: affected pages turn red. | | Size | Inlinks, Link score, Same | The size of each dot. | | Levels | All, 1 to 6 | How deep the diagram goes. Deeper pages fold into a count on their ancestor. | | All internal links | On or off | Draws every internal link, faintly, under the tree. Not in the directory tree. | | Find a page on the map | Text | Rings the pages whose address or title contains it. Enter zooms to them. | | Colour by | Legend | | --- | --- | | Topic | The topics, the first twelve. | | Status | 2xx, 3xx, 4xx / 5xx | | Indexability | Indexable, Non-indexable, Bot check | | Crawl depth | One colour per level from 0, the last for that level and deeper. | | Response time | < 400 ms, 400–800, 800–1500, > 1500 | | A check | The check's name, and **not affected**. | The mouse: scroll to zoom, drag to move, hover for a page's title, address, status, depth and inlinks. A click on a page opens its card; a double-click shows only its branch. The toolbar's **Export** saves the pages by topic. ## Page card | Part | What it shows | | --- | --- | | Title and address | The page's title, or its address; a click on the address opens it in the table. | | Figures | Status, Depth, Inlinks, Words, Topic, Indexability. | | Below it | **N directly below**, and **N folded** when levels are cut. | | Checks | Up to eight checks the page fails, most important first. A check opens in [Issues](https://www.getlooploop.com/gllmolfar/docs/issues). With none: **No checks flag this page.** | | Buttons | **In table**, **About the page**, **Only this branch**, **Whole site** (after a branch was picked) and **Open in browser**. | ## States and messages - **No crawl open** — the app says: "Enter a site address above and press Start." Nothing to draw yet. - **Crawl running** — the app says: "The structure is drawn when the crawl ends." The click paths are known only once every page is in. - **Nothing clicked** — the app says: "Click a page on the diagram to see it here: its numbers, its checks, and what to do next." The card before a page is picked. - **Image saved** — the app says: "Image saved to {path}" After **PNG**: the diagram as it is on screen, on a white background. > **Locked tab.** A tab with a lock is not open on this account; pressing it says what opens it. ## Related pages - [All pages (Internal)](https://www.getlooploop.com/gllmolfar/docs/all-pages.md): The All pages tab, called Internal in Pro mode: every crawled URL in one table, with filters, search, column sets, an overview panel and a page inspector. - [Topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map.md): The Topic map tab: every embedded page as a dot, placed by meaning and coloured by topic, status, indexability or depth, with a card for each page. - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. --- # URL audit > One page, no crawl. GLLmolfar fetches the address, renders it with JavaScript, photographs it on desktop and mobile, and runs the page's own checks. Checks that need the rest of the site are left out. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/url-audit - **Where:** Site › URL audit - **Mode:** Simple and Pro - **Built from:** One fetch of the address, from this computer - **Changes:** Each time you press Audit ## On the screen At the top, the address field. After an audit: the address with four actions, a row of tiles, the page map, and two columns — the checks and findings on the left, the screenshots and headers on the right. ![The URL audit tab of GLLmolfar after an audit of the getlooploop.com home page](https://www.getlooploop.com/images/gllmolfar/docs/url-audit.webp) Numbered parts of the screenshot: 1. **Page address.** Type an address and press **Audit**. The line under it says what the audit does and what it leaves to a crawl. 2. **Actions.** **Open in browser**, **Show in the crawl** (only when the open crawl has the page), **Copy HTML** and **Reverse prompts**. 3. **Tiles.** Status, indexability, response time, HTML size and what JavaScript changes. A **Bot check** tile joins them when the page answered with one. 4. **Page map.** The rendered page cut into its blocks and coloured four ways. Built on a button, for desktop or mobile. 5. **Checks this page fails.** The page-level checks it fails, most important first, each with how to fix it. 6. **Screenshots.** The page on desktop and as **Mobile (Googlebot smartphone)**. 7. **Response headers.** The headers that matter, security headers as ticks and crosses, cookies, and all headers in a fold. ## Tiles | Tile | What it shows | | --- | --- | | Status | The HTTP status and its text. Red from 400 or with no answer, amber for a redirect. | | Indexability | **Indexable** or **Non-indexable**, with the reason when it is not. | | Response | The server's response time in ms, amber over 1,000 ms. Under it: **audit took N s**. | | HTML size | The HTML in KB and the number of words. | | JavaScript | **Changes the page** with what changed (canonical, meta robots, title, H1, description, added links), or **No key changes** with words in the HTML against words after JavaScript. Red when JavaScript changes the canonical or meta robots. | | Bot check | Shown only when the page answered with a CAPTCHA or bot check, with its name. | ## Page map The page as the built-in browser renders it, cut into its blocks. **Build the page map** draws it; **Desktop** and **Mobile (Googlebot smartphone)** switch the device. Once built, four buttons pick what the colours show, and a list beside the picture names the blocks. A block in the list scrolls the picture to it. | Colours | What they show | | --- | --- | | Meaning to a question | Each block coloured by how close its text is to a question, from this page's lowest to its highest. The first question is the page's H1, or its title; type another and press **Colour**. Beside it: **Closest blocks**. Needs embeddings switched on. | | What AI quotes | What an AI retriever takes from the page, in reading order, the first 26 passages kept. Blocks read **quoted**, **weak passage**, **past the 26 passages**, **only after JavaScript** or **menu / footer**, with a count of each and **What to do** for the ones that miss. | | Problems | Accessibility findings counted inside each block: fields without a label, buttons and links without a name, images without alt. | | Attention forecast | Heat from the layout: position against the first screen, size, headings, calls to action and large images. A line marks the end of the first screen. Lists **Most likely seen first** and **Content that gets little attention**. | > **A forecast, not behaviour.** The attention view says so on screen: “A forecast from layout, not people's behaviour. Real clicks and scrolling need an analytics script on the site.” ## The two columns | Block | What it shows | | --- | --- | | Checks this page fails | The crawl's page checks that this page fails, rated like the Issues tab, most important first, each with how to fix it. Checks that need the rest of the site — links in, sitemap, duplicates, topics — are left out. | | Accessibility (axe) | axe-core run in the rendered page: rules broken by impact (Critical, Serious, Moderate, Minor), the elements affected, the WCAG references, and **How to fix** for each rule. The note under it gives the axe-core version, the rules passed and the ones that need a person to look. | | In search results | The title and description drawn as a desktop search result, with their widths in pixels against the limit. | | Redirects | Each hop with its status, ending at the final address. Shown only when the address redirected. | | What JavaScript changes | Title, canonical, meta robots, H1 and description in the HTML against after JavaScript, links added, and the HTML size before and after rendering. | | robots.txt | Whether Googlebot, GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot and Google-Extended may fetch this address, with the matching rule and its line. | | Screenshots | Desktop, and **Mobile (Googlebot smartphone)**. | | Response headers | content-type, cache-control, content-encoding, x-robots-tag, last-modified, etag, vary, server and link, when sent. Then HSTS, CSP, X-Content-Type-Options, X-Frame-Options and Referrer-Policy as ✓ or ✕, and **No compression** when there is none. **Cookies set by this page: N** lists each cookie as Secure or **not Secure**, HttpOnly or **readable by scripts**, and its SameSite. **All headers (N)** opens the rest. | ## States and messages - **Running** — the app says: "Fetching, rendering and taking screenshots. Usually 5–20 seconds." Under the field, a bar with the seconds so far and, once an audit has run on this computer, how long one usually takes. The button reads **Checking…** until the audit ends. - **No screenshot** — the app says: "The page did not load in time for a picture." Shown in place of a screenshot. - **Not rendered** — the app says: "The page could not be rendered." What JavaScript changes cannot be compared. - **No accessibility check** — the app says: "The accessibility check did not run: the page did not load in a browser window." The Accessibility block stays empty. - **No robots.txt** — the app says: "No robots.txt on this site: everything is allowed." The robots.txt block has no table. - **Page map failed** — the app says: "The page map could not be built." The rest of the audit stays on screen. - **Failed** — the app says: "The audit could not run." Or **The audit stopped:** with the reason. > **Locked tabs.** A tab with a lock is not open on this account; pressing it says what opens it. ## Related pages - [Heat map](https://www.getlooploop.com/gllmolfar/docs/heat-map.md): Site › Heat map: the crawl's folders as tiles sized by pages and coloured by pages without problems, problems per page, Google clicks or lost clicks. - [Reverse prompts](https://www.getlooploop.com/gllmolfar/docs/reverse-prompts.md): Site › Reverse prompts: GLLmolfar reads a page into blocks, your AI agent describes what each does, and every block gets a reusable prompt and an example. - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. --- # Heat map > The crawled site as tiles, one per folder. A tile's size is the number of pages in it; its colour is the metric you pick. The side panel names the folders where it is hottest. A tile opens its pages in the table. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/heat-map - **Where:** Site › Heat map - **Mode:** Simple and Pro - **Built from:** The open crawl, and Search Console when connected - **Changes:** When you open another crawl or pick a metric ## On the screen The heading counts the pages on the map. Under it, the map; on the right, the metric's scale and the folders where it is hottest. ![The Heat map tab of GLLmolfar for a crawl of getlooploop.com, coloured by pages without problems](https://www.getlooploop.com/images/gllmolfar/docs/heat-map.webp) Numbered parts of the screenshot: 1. **Heading.** **N pages by folder**, and how to read the map. 2. **Colour by.** **Pages without problems**, **Problems per page**, **Google clicks**, **Lost clicks**. The last two need Search Console data for the site. 3. **Folders.** Each first-level folder is a frame with its name and page count; the tiles inside are its subfolders, with the value of the metric. 4. **Scale.** What the metric means, and its colours from **worse** to **better**. 5. **Where it is hottest.** Six subfolders where the metric is worst, with the value and the number of pages. A row opens its pages. 6. **Other.** Folders past the twelve largest, drawn together as one frame. ## Colour by | Metric | What a tile's value is | | --- | --- | | Pages without problems | Share of a folder's pages with no serious finding. The scale runs from red (worse) to green (better). | | Problems per page | Average number of findings per page in the folder: the crawl's failed checks and each kind of accessibility finding. | | Google clicks | Search Console clicks of the folder's pages, last 28 days. From the same report as [Search Console › Performance](https://www.getlooploop.com/gllmolfar/docs/search-console). | | Lost clicks | Clicks the folder's pages lost against the 28 days before, page by page as in [Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks). | > **Without Search Console.** **Google clicks** and **Lost clicks** are greyed out while there is no Search Console data for this site; hovering says “Connect Search Console for this site to colour by clicks.” A report saved on this computer counts: signed out or offline, the map colours by the clicks of the last refresh. See [Search Console](https://www.getlooploop.com/gllmolfar/docs/search-console). ## How the tiles are grouped - Only HTML pages that answered 200 are on the map. - Pages are grouped by the first segment of their address, then by the second. The home page is the `/` frame. - The twelve largest folders get a frame each; the rest are drawn together as **Other**. - Inside a frame, the twelve largest subfolders get a tile each; the rest share one tile labelled **+N more**. - Small tiles show no text. Hovering any tile gives its folder, its pages and its value. - A tile, or a row of **Where it is hottest**, opens its pages in the table. ## States and messages - **No crawl open** — the app says: "Crawl a site to see its heat map." The tab needs the pages of a crawl. - **No Search Console** — the app says: "Connect Search Console for this site to colour by clicks." No Search Console data for this site, live or saved: the two click metrics cannot be picked. ## Related pages - [URL audit](https://www.getlooploop.com/gllmolfar/docs/url-audit.md): Site › URL audit: one page checked without a crawl — status, speed, JavaScript, page map, failed checks, accessibility, robots.txt, headers and screenshots. - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks.md): GSC › Lost clicks: pages that lost Google clicks over the last 28 days, the reason the numbers point at, the lost queries, and what the crawl found there. - [Structure](https://www.getlooploop.com/gllmolfar/docs/structure.md): The Structure tab: the crawled site drawn as a crawl tree, a directory tree or a force-directed diagram, coloured by topic, status, depth, speed or a check. - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. --- # Search Console: Performance > Google Search Console for the open site, read-only: clicks, impressions, CTR and position by day, this week's query moves, and the queries, pages, countries and devices behind them. Performance is the first of the nine reports in the GSC section, and the place where you connect. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/search-console - **Where:** GSC › Performance - **Mode:** Simple and Pro - **Built from:** Your Search Console property, read from Google and kept on this computer - **Changes:** On Refresh, or when you pick another property or period ## On the screen GSC in the rail opens a panel of nine reports: Performance, Pages, Positions, Indexing, Sitemaps, Lost clicks, Cannibalization, Query clusters and Visualisation. Every report shares the same head: what the section reads, the property, and the period. The picture shows a saved report while signed out. ![The Performance report of the GSC section in GLLmolfar, a saved report for getlooploop.com shown while signed out](https://www.getlooploop.com/images/gllmolfar/docs/search-console.webp) Numbered parts of the screenshot: 1. **The GSC section.** Nine reports, one page each. **Lost clicks** and **Cannibalization** carry a count in the panel when they have rows. 2. **What it reads.** Clicks, impressions, CTR and position for the last 28 days, and the queries and pages behind them. Read-only; the data comes from Google straight to this computer. 3. **Saved report.** **Search Console data from {date}.** and why it is not fresh, with the one button that refreshes it: here **Sign in with Google**. 4. **Property and period.** The property the report comes from, and **Last 7 days**, **Last 28 days** or **Last 3 months**. 5. **Totals.** **Total clicks**, **Total impressions**, **Average CTR**, **Average position**, each against the period before. A ticked card draws its line on the chart. 6. **By day.** The ticked metrics by day, each on its own scale; the first two have an axis, left and right. 7. **Queries this week.** This week against last week: **new queries**, **lost queries**, **rose in position**, **fell in position**. ## Connecting - Signed out, with no saved report, the section says “Search Console connects to the Google account you sign in to GLLmolfar with.” beside **Sign in with Google**. - Signed in, **Connect Search Console** opens Google in your browser. The section reads “Waiting for Google in your browser…” with **Open the page again**, **Copy link** and **Cancel**. - Use the same Google account you are signed in to GLLmolfar with. Another account is refused with a message. - Once connected, the head shows **Search Console property**, the period, **Refresh**, **Connected as {email}.** and **Disconnect**. - The property that covers the open crawl's site is picked for you. When none covers it, the section says so and you pick one from the list. > **Read-only.** “Google asks for read-only access; GLLmolfar can change nothing in your Search Console.” Every report in the section reads; none of them writes to Google. ## Kept on this computer A report, once read, is saved on this computer with its date. Offline, signed out or disconnected, the last saved report is shown in full under a line that says how old it is, and every report in the section uses it. Only refreshing needs the account. | Line over the report | What refreshes it | | --- | --- | | Search Console data from {date}. Sign in to refresh it; until then this is the last saved report. | **Sign in with Google** | | Search Console data from {date}. Connect Search Console to refresh it; until then this is the last saved report. | **Connect Search Console** | | Search Console data from {date}. No connection to Google right now; this is the last saved report. | **Try again** | | Search Console data from {date}. Google did not answer the refresh; this is the last saved report. | **Try again** | > **What the app says about access and data.** “Our server keeps only the encrypted access token, to renew it; the reports stay on this computer. Signing out deletes the token.” Signing out or disconnecting forgets the token, not the reports. ## Performance A line under the title gives the dates against the period before, says Search Console data trails by about two days, and when the report was read. | Block | What it shows | | --- | --- | | Totals | **Total clicks**, **Total impressions**, **Average CTR**, **Average position**, each with the change against the period before: percent for clicks and impressions, points (pp) for CTR, places for position. A card switches its line on the chart on and off. | | Chart | The chosen metrics by day, each on its own scale; position runs upside down, 1 at the top. Hovering shows every value of that day. | | Queries this week | The last seven days against the seven before. A query moved in position when it had at least 3 impressions in both weeks and changed by a whole place or more. A card opens its list: query, **Position before**, **Position now**, impressions, clicks. | | Tables | **Queries**, **Pages**, **Countries**, **Devices**, each with clicks, impressions, CTR and average position. **Filter…** narrows the rows; **Export** saves them. 100 rows are shown, with **Showing N of M.** | > **Clicks not in the list.** When queries Google keeps private bring clicks, a line says how many: they are in the totals but not in the query list. Under the Queries and Pages tables, a line says how many rows Google gave by clicks; the export has them all. ## The other reports - [Pages](https://www.getlooploop.com/gllmolfar/docs/gsc-pages): each page with its top query and its state in the crawl, and the crawled pages Google never showed. - [Positions](https://www.getlooploop.com/gllmolfar/docs/gsc-positions): queries by position band, and those close to the top. - [Indexing](https://www.getlooploop.com/gllmolfar/docs/gsc-indexing): Google's URL Inspection answer, page by page. - [Sitemaps](https://www.getlooploop.com/gllmolfar/docs/gsc-sitemaps): the sitemaps submitted to Search Console and how Google read them. - [Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks): pages that lost clicks, and the reason the numbers point at. - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization): queries for which Google showed two or more of your pages. - [Query clusters](https://www.getlooploop.com/gllmolfar/docs/gsc-query-clusters): queries grouped by meaning on this computer. - [Visualisation](https://www.getlooploop.com/gllmolfar/docs/gsc-visualisation): the period's queries drawn instead of listed. ## Where else it shows - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview): a Search Console card with the last 28 days and this week's moves, and **Lost in Google** beside the fixes. - [AI](https://www.getlooploop.com/gllmolfar/docs/ai): **Questions people search**, the queries phrased as questions, ready to check against the site's passages. - [Heat map](https://www.getlooploop.com/gllmolfar/docs/heat-map): **Google clicks** and **Lost clicks** as colours. ## States and messages - **Signed out** — the app says: "Search Console connects to the Google account you sign in to GLLmolfar with." With **Sign in with Google**, when no report has been saved yet. - **Reading** — the app says: "Reading Search Console: {step} — step {n} of {total}" The wait bar moves through the steps: totals, clicks by day, queries, pages, countries, devices, pages by query. - **No properties** — the app says: "This Google account has no Search Console properties." Nothing to read. - **Wrong account** — the app says: "That was a different Google account. Connect Search Console with the account you are signed in to GLLmolfar with." Connect again with the right account. - **Box not ticked** — the app says: "Google was not given access to Search Console. Connect again and tick the Search Console box." The connection was made without Search Console access. - **Disconnected** — the app says: "Search Console is not connected any more (access was removed in the Google account). Connect it again." Access was taken back on Google's side. - **Slow down** — the app says: "Google asks to slow down: the allowance for this site is used up for now." Google's own limit; try again later. - **No data** — the app says: "No search data for this period." The chart is empty for the period picked. > **Locked reports.** A report with a lock is not open on this account; pressing it says what opens it. ## Related pages - [Search Console: Pages](https://www.getlooploop.com/gllmolfar/docs/gsc-pages.md): GSC › Pages: each page's Google clicks, impressions, CTR and position, its top query, its state in the crawl, and crawled pages Google never showed. - [Search Console: Indexing](https://www.getlooploop.com/gllmolfar/docs/gsc-indexing.md): GSC › Indexing: Google's URL Inspection answer page by page: in the index or not and why, last crawl, chosen canonical, rich results, kept on this computer. - [Search Console: Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks.md): GSC › Lost clicks: pages that lost Google clicks over the last 28 days, the reason the numbers point at, the lost queries, and what the crawl found there. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [Heat map](https://www.getlooploop.com/gllmolfar/docs/heat-map.md): Site › Heat map: the crawl's folders as tiles sized by pages and coloured by pages without problems, problems per page, Google clicks or lost clicks. - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. --- # Search Console: Pages > The pages Google showed in the period, with their clicks, impressions, CTR and position, the query that brought the most clicks, and what this crawl found at each address. Under them, the crawl's indexable pages that Google did not show at all. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-pages - **Where:** GSC › Pages - **Mode:** Simple and Pro - **Built from:** The saved Search Console report and the open crawl - **Changes:** With the period, on Refresh, and with the crawl ## On the screen The head is the one every report shares: the property, the period and, when the report is a saved one, its date (see [Performance](https://www.getlooploop.com/gllmolfar/docs/search-console)). Below it, the table of pages, most clicks first. ![The Pages report of the GSC section in GLLmolfar for getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/gsc-pages.webp) Numbered parts of the screenshot: 1. **Period.** The dates against the period before, and when the report was read. 2. **Filter….** Narrows the rows to addresses that contain the text. 3. **Address.** The page, shortened to its path. When the crawl has the page, it opens its details. 4. **Clicks to Average position.** Clicks, impressions, CTR and average position for the period. 5. **Top query.** The query that brought the page the most clicks; **+N** is how many more queries it was shown for. 6. **In the crawl.** **indexable**, the reason the crawl found it is not, or **not in this crawl**. ## The table | Column | What it shows | | --- | --- | | Address | The page Google showed. 100 rows are shown, with **Showing N of M.** | | Clicks, Impressions, CTR, Average position | Search Console's figures for the page in the period picked. | | Top query | Read from the page-by-query rows of the last 28 days: the query with the most clicks, then the most impressions. Empty when the page has no listed query. | | In the crawl | **indexable** in green; an error status, a redirect or the reason the page is not indexable, in red; **not in this crawl** when the open crawl did not reach the address. | ## Crawled, never shown in Google Under the table, **Crawled, never shown in Google: {n}** lists the crawl's indexable HTML pages that had no impression in the period. A line says how many of all indexable pages that is: “Often new, orphaned or too thin for any query.” Each row has the address, inlinks, depth and words. - **No crawl** — the app says: "Open a crawl of this site to see which of its pages Google never showed." The list needs the crawl to compare with. - **Limit reached** — the app says: "Search Console returned its limit of {n} pages, so a page missing from it may still have impressions; this list is not drawn." Google's row limit makes the comparison unsafe, so it is left out. ## Related pages - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Indexing](https://www.getlooploop.com/gllmolfar/docs/gsc-indexing.md): GSC › Indexing: Google's URL Inspection answer page by page: in the index or not and why, last crawl, chosen canonical, rich results, kept on this computer. - [Search Console: Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks.md): GSC › Lost clicks: pages that lost Google clicks over the last 28 days, the reason the numbers point at, the lost queries, and what the crawl found there. - [All pages (Internal)](https://www.getlooploop.com/gllmolfar/docs/all-pages.md): The All pages tab, called Internal in Pro mode: every crawled URL in one table, with filters, search, column sets, an overview panel and a page inspector. - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. --- # Search Console: Positions > Where the site's queries stand in Google: every query once, in the band of its average position, and the list of queries close to the top, the ones on the lower half of page one and on page two. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-positions - **Where:** GSC › Positions - **Mode:** Simple and Pro - **Built from:** The queries of the saved Search Console report - **Changes:** With the period, and on Refresh ## On the screen Under the shared head (property, period, saved-report line: see [Performance](https://www.getlooploop.com/gllmolfar/docs/search-console)), two blocks: the bands, then the queries close to the top. ![The Positions report of the GSC section in GLLmolfar for getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/gsc-positions.webp) Numbered parts of the screenshot: 1. **Period.** The dates against the period before, and when the report was read. 2. **Queries by average position.** How many queries the period has. Each query once, in the band of its average position. 3. **Bands.** **1–3**, **4–10**, **11–20**, **21–50**, **51+**, each bar as long as its number of queries against the largest band. 4. **Band figures.** The queries in the band, and their clicks and impressions together. 5. **Close to the top.** Queries with an average position from 4 to 20, most impressions first. 6. **The list.** **Query**, **Average position**, **Impressions**, **Clicks**, **CTR**. ## The two blocks | Block | What it shows | | --- | --- | | Queries by average position: {n} | Every query of the period in one of five bands, by its average position. Beside each bar: the number of queries, then **{clicks} clicks · {imp} impressions**. | | Close to the top: {n} | “Average position 4 to 20 — the lower half of page one and page two — with the most impressions first. A small improvement here moves clicks most.” Up to 100 queries. | The bands are counted from the queries Search Console lists; queries Google keeps private are not among them. The same five bands are drawn on a log scale in [Visualisation](https://www.getlooploop.com/gllmolfar/docs/gsc-visualisation). ## States and messages - **None near the top** — the app says: "No query sits between positions 4 and 20 in this period." Shown in place of the list. ## Related pages - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Visualisation](https://www.getlooploop.com/gllmolfar/docs/gsc-visualisation.md): GSC › Visualisation: the period's queries drawn: words as a treemap, spread by position, CTR and impressions, opportunities, click concentration, brand. - [Search Console: Query clusters](https://www.getlooploop.com/gllmolfar/docs/gsc-query-clusters.md): GSC › Query clusters: up to 1000 Search Console queries grouped by meaning on this computer, with each cluster's clicks, position, pages and queries. - [Search Console: Pages](https://www.getlooploop.com/gllmolfar/docs/gsc-pages.md): GSC › Pages: each page's Google clicks, impressions, CTR and position, its top query, its state in the crawl, and crawled pages Google never showed. --- # Search Console: Indexing > Google's own answer for each page of the site, from Search Console's URL Inspection: whether the page is in the index and, if not, why; when Google last crawled it; which canonical Google chose; and which rich results it found. Pages are asked in portions, and every answer is kept on this computer. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-indexing - **Where:** GSC › Indexing - **Mode:** Simple and Pro - **Built from:** Search Console's URL Inspection, asked page by page; the crawl's indexable pages - **Changes:** When you press Inspect, a portion at a time ## On the screen Under the shared head (see [Performance](https://www.getlooploop.com/gllmolfar/docs/search-console)): the lead, the button with the day's allowance, four counts, the reasons in Google's words, and one row per page asked. ![The Indexing report of the GSC section in GLLmolfar for getlooploop.com, fifteen pages inspected](https://www.getlooploop.com/images/gllmolfar/docs/gsc-indexing.webp) Numbered parts of the screenshot: 1. **What it asks.** In the index or not and why, when Google last fetched the page, which canonical it chose. Every answer is kept on this computer. 2. **Inspect N pages.** Asks Google about the next portion: up to 200 pages not yet asked or older than 14 days. 3. **Allowance.** **{left} of {daily} inspections left today**, and how many pages are not yet asked or older than 14 days. 4. **Counts.** **indexed**, **not indexed or excluded**, **Google chose another canonical**, **not asked yet**. 5. **By Google's reason.** The pages asked, grouped by the reason Search Console gives, in its own words. 6. **In the index.** **Indexed**, **Not indexed**, **Excluded**, or **No answer**. 7. **Canonical.** **as declared**, or **Google chose {address}** in red when Google picked another page. ## How pages are asked - The pages asked about are the open crawl's indexable HTML pages that answered 200. Without a crawl, the pages Search Console lists for the period. - **Inspect {n} pages** takes up to 200 of them at a press: pages never asked, or asked more than 14 days ago. - Google is asked four pages at a time. While it runs, the wait bar reads “Asking Google about page {n} of {total}” beside **Stop**; what was learnt before Stop is kept. - Search Console allows 2,000 page inspections a day per site; that is Google's limit, not the app's. The app counts the day's use and stops when it is used up. - Each answer is saved on this computer with its date, so the report is there offline and signed out. Asking again needs the account. ## The table | Column | What it shows | | --- | --- | | Address | The page asked about. Pages not in the index come first, then those with no answer, then indexed pages. Up to 500 rows, with **Showing N of M.** | | In the index | Google's verdict: **Indexed**, **Not indexed**, **Excluded** (neither), or **No answer** when the question failed. | | Google's reason | The coverage state as Search Console states it, such as “Submitted and indexed”. | | Last crawled by Google | When Google last fetched the page, or a dash. | | Canonical | **as declared** when Google kept the page's own canonical; **Google chose {address}** when it picked another. | | Rich results | The rich result types Google detected, such as Breadcrumbs, or a dash. | ## States and messages - **Nothing asked** — the app says: "No page has been inspected yet. Inspect a portion to see which pages Google indexed and why the others are out." Shown in place of the counts. - **Running** — the app says: "Asking Google about page {n} of {total}" With **Stop**. - **Allowance used** — the app says: "Google's daily allowance of inspections for this site is used up; it renews tomorrow." The answers so far are kept. ## Related pages - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Sitemaps](https://www.getlooploop.com/gllmolfar/docs/gsc-sitemaps.md): GSC › Sitemaps: the sitemaps submitted to Search Console, their URLs, errors and warnings, when Google last read them, beside the URLs the crawl found. - [Search Console: Pages](https://www.getlooploop.com/gllmolfar/docs/gsc-pages.md): GSC › Pages: each page's Google clicks, impressions, CTR and position, its top query, its state in the crawl, and crawled pages Google never showed. - [Issues](https://www.getlooploop.com/gllmolfar/docs/issues.md): The Issues tab: every check that failed in the crawl, by importance and type, with why it matters, how to fix it and the URLs behind it. - [URL audit](https://www.getlooploop.com/gllmolfar/docs/url-audit.md): Site › URL audit: one page checked without a crawl — status, speed, JavaScript, page map, failed checks, accessibility, robots.txt, headers and screenshots. --- # Search Console: Sitemaps > The sitemaps submitted for the property in Search Console and how Google read them: how many URLs each submits, when it was submitted and last read by Google, and its errors and warnings. A line compares them with the URLs this crawl found in the site's sitemaps. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-sitemaps - **Where:** GSC › Sitemaps - **Mode:** Simple and Pro - **Built from:** Search Console's sitemaps list, kept on this computer; the open crawl - **Changes:** On this report's Refresh; a saved list is used for a day ## On the screen Under the shared head (see [Performance](https://www.getlooploop.com/gllmolfar/docs/search-console)): the report's own **Refresh**, four counts, the comparison with the crawl, and one row per sitemap. ![The Sitemaps report of the GSC section in GLLmolfar for getlooploop.com, one sitemap](https://www.getlooploop.com/images/gllmolfar/docs/gsc-sitemaps.webp) Numbered parts of the screenshot: 1. **Refresh.** Reads the list from Google again, and when it was read. It needs the account, so it is off on a saved report. 2. **Counts.** **sitemaps**, **URLs submitted**, **Errors** (red when there are any) and **Warnings**, over all the sitemaps. 3. **Against the crawl.** **This crawl found {n} URLs in the site's sitemaps.** 4. **Sitemap and Type.** The sitemap's address, and **sitemap**, **sitemap index** or the type Search Console gives. 5. **URLs submitted, Submitted.** How many URLs the sitemap submits, and when it was submitted. 6. **Last read by Google.** When Google last downloaded it, then its errors and warnings. ## The table | Column | What it shows | | --- | --- | | Sitemap | The address of the sitemap. **waiting for Google** beside it when Search Console has not processed it yet. | | Type | **sitemap index** for an index of sitemaps; otherwise the type Search Console gives, such as sitemap. | | URLs submitted | The URLs Search Console counts in it. | | Submitted, Last read by Google | When it was submitted, and when Google last downloaded it; a dash when unknown. | | Errors, Warnings | What Search Console reported when it read the sitemap. Errors are red. | The list is saved on this computer and reused for a day; **Refresh** reads it again. The sitemaps the crawl itself read are under [Structure](https://www.getlooploop.com/gllmolfar/docs/structure) and [Crawling settings](https://www.getlooploop.com/gllmolfar/docs/settings-crawling). ## States and messages - **Reading** — the app says: "Reading the sitemaps submitted to Search Console" The wait bar while the list is read. - **None submitted** — the app says: "No sitemap has been submitted for this property in Search Console." Search Console has no sitemap for the property. ## Related pages - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Indexing](https://www.getlooploop.com/gllmolfar/docs/gsc-indexing.md): GSC › Indexing: Google's URL Inspection answer page by page: in the index or not and why, last crawl, chosen canonical, rich results, kept on this computer. - [Structure](https://www.getlooploop.com/gllmolfar/docs/structure.md): The Structure tab: the crawled site drawn as a crawl tree, a directory tree or a force-directed diagram, coloured by topic, status, depth, speed or a check. - [Settings › Crawling](https://www.getlooploop.com/gllmolfar/docs/settings-crawling.md): Settings › Crawling: request speed, connections, the user agent, the timeout, robots.txt, a pause at a CAPTCHA, JavaScript rendering and service pages. --- # Search Console: Lost clicks > Pages that lost a noticeable share of their Google clicks, the last 28 days against the 28 before. Each comes with the reason the numbers point at, the queries it lost, and, where the crawl has one, the crawl's own reason. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks - **Where:** GSC › Lost clicks - **Mode:** Simple and Pro - **Built from:** The saved Search Console report (last 28 days and the 28 before) and the open crawl - **Changes:** On Refresh ## On the screen The picture shows the empty state: in this period no page of getlooploop.com lost enough clicks to be listed. With losses, the panel item carries their count, and the report shows a row of reason chips and a table. ![The Lost clicks report of the GSC section in GLLmolfar for getlooploop.com, with no page listed](https://www.getlooploop.com/images/gllmolfar/docs/gsc-losses.webp) Numbered parts of the screenshot: 1. **Lost clicks in the panel.** Carries the number of pages listed, when there are any. 2. **The two periods.** The 28 days compared with the 28 before, whatever period the head shows. 3. **Nothing lost.** **No page lost a noticeable share of its clicks.** ## Which pages are listed “A page is listed when it had at least 10 clicks before and lost 30% of them, and at least 5.” The line under the table says so, with **Showing N of M.** (up to 100 rows). ## With losses | Block | What it shows | | --- | --- | | Reason chips | One chip per reason with its number of pages, most first, and **Export**, which saves every row. | | Address | The page. Under it, **Lost queries**: up to three, each with its position before → now, or **gone**. | | Reason | The reason the numbers point at, and beside it what this crawl found at the address. | | Clicks, Impressions, Position, CTR | Before → after for each, with the change in clicks. | | Reason | What it means | | --- | --- | | Gone from search | No impressions in the last 28 days: the page left Google's results — removed, redirected, noindex, or no longer ranked. | | Position fell | The average position is two places or more worse: another page — yours or a competitor's — took the place. | | Lower CTR | Impressions held, clicks did not: the snippet, or what now stands around it in the results (ads, AI answers, rich results), draws fewer clicks. | | Less demand | Impressions fell 30% or more at the same position: fewer people search for it — the season, or the topic. | | Several causes | Clicks fell with no single cause in the numbers: a little position, a little demand, a little CTR. | | From the crawl | An error status, **Redirected**, the reason the page is not indexable, or **Not found by the crawl**. Nothing when the crawl found the page indexable. | The same losses appear as **Lost in Google** on the [Overview](https://www.getlooploop.com/gllmolfar/docs/overview) and as **Lost clicks** on the [Heat map](https://www.getlooploop.com/gllmolfar/docs/heat-map). ## States and messages - **None** — the app says: "No page lost a noticeable share of its clicks." No page passes the rule above. - **Old report** — the app says: "This report was fetched before losses and cannibalization were added. Press Refresh to fetch it again." A report saved by an older version. ## Related pages - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Pages](https://www.getlooploop.com/gllmolfar/docs/gsc-pages.md): GSC › Pages: each page's Google clicks, impressions, CTR and position, its top query, its state in the crawl, and crawled pages Google never showed. - [Search Console: Cannibalization](https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization.md): GSC › Cannibalization: queries for which Google showed two or more of your pages, with each page's clicks, share of impressions and position. - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [Heat map](https://www.getlooploop.com/gllmolfar/docs/heat-map.md): Site › Heat map: the crawl's folders as tiles sized by pages and coloured by pages without problems, problems per page, Google clicks or lost clicks. --- # Search Console: Cannibalization > Queries for which Google showed two or more of the site's pages in the last 28 days. The page with the most clicks comes first and is marked **keep**; the others compete with it. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization - **Where:** GSC › Cannibalization - **Mode:** Simple and Pro - **Built from:** The page-by-query rows of the saved Search Console report, last 28 days - **Changes:** On Refresh ## On the screen The picture shows the empty state: in this period no query of getlooploop.com was split between pages. With some, the panel item carries their count and the report lists one block per query. ![The Cannibalization report of the GSC section in GLLmolfar for getlooploop.com, with no query listed](https://www.getlooploop.com/images/gllmolfar/docs/gsc-cannibals.webp) Numbered parts of the screenshot: 1. **Cannibalization in the panel.** Carries the number of queries listed, when there are any. 2. **The term.** The dotted underline explains the word in one line, with a link to the GetLoopLoop glossary. 3. **What it lists.** Queries for which Google showed two or more of your pages, and the dates. 4. **Nothing split.** **No query is split between pages.** ## Which queries are listed “A page counts for a query when it has at least 10 impressions and 10% of the query's impressions.” A query is listed when two or more pages count for it. ## With queries | Element | What it shows | | --- | --- | | Filter… and Export | **Filter…** narrows to queries or pages that contain the text; **Export** saves every query with its pages. | | Query | The query, and **{pages} pages · {impressions} impressions · {clicks} clicks**. | | Pages | One row per page, most clicks first: **keep** on the first, then its clicks, its impressions with its share of the query's impressions, and its average position. | | Under the list | **Showing N of M.** (up to 100 queries) and the rule above. | > **Not the same as the crawl's cannibalization.** This report reads what Google showed for real queries. [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization) under Meaning compares the pages' texts by meaning, from the crawl. A query split here between pages that are also a pair there is a strong case to resolve. ## States and messages - **None** — the app says: "No query is split between pages." No query passes the rule above. - **Limit reached** — the app says: "Search Console returned {n} page–query rows, its limit: the rarest pairs are not included." Added to the line under the title. - **Old report** — the app says: "This report was fetched before losses and cannibalization were added. Press Refresh to fetch it again." A report saved by an older version. ## Related pages - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization.md): The Cannibalization tab: groups of pages that say almost the same thing, the page to keep, the pages side by side and their text passage by passage. - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Query clusters](https://www.getlooploop.com/gllmolfar/docs/gsc-query-clusters.md): GSC › Query clusters: up to 1000 Search Console queries grouped by meaning on this computer, with each cluster's clicks, position, pages and queries. - [Search Console: Lost clicks](https://www.getlooploop.com/gllmolfar/docs/gsc-lost-clicks.md): GSC › Lost clicks: pages that lost Google clicks over the last 28 days, the reason the numbers point at, the lost queries, and what the crawl found there. - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. --- # Search Console: Query clusters > Up to 1000 queries of the last 28 days grouped by meaning, on request, by the embedding model on this computer. Each cluster has its clicks, impressions and position, its words, the pages that get its clicks and its queries. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-query-clusters - **Where:** GSC › Query clusters - **Mode:** Simple and Pro - **Built from:** The saved report's queries, embedded on this computer - **Changes:** On Cluster the queries; made once per report ## On the screen Before the first run, the report has the note and **Cluster the queries**. After it, the clusters, most clicks first; a cluster opens to show its words, pages and queries. ![The Query clusters report of the GSC section in GLLmolfar, 23 clusters from 231 queries of getlooploop.com](https://www.getlooploop.com/images/gllmolfar/docs/gsc-clusters.webp) Numbered parts of the screenshot: 1. **On this computer.** Grouped by the embedding model on this computer. Nothing is sent anywhere. It takes about a minute. 2. **Count.** **{n} clusters from {q} queries**. 3. **Export.** Saves every cluster with its queries. 4. **Cluster.** Named by its words; the row opens the cluster. A name may be a competitor's, replaced in this picture. 5. **Share of clicks.** The bar is the cluster's share of the clicks of all clusters. 6. **Figures.** **{clicks} clicks · {impressions} impressions · pos. {position}**, then the number of queries. 7. **Split between pages.** Two or more pages each take a real share of this cluster's clicks: one topic, several pages. ## Making the clusters - Nothing runs on its own: **Cluster the queries** starts it, and the button reads **Clustering…** until it ends. - It takes the 1000 queries of the last 28 days with the most impressions, whatever period the head shows, and needs at least six. - The model is the app's embedding model on this computer (see [Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings)); the queries do not leave it. - The clusters are saved on this computer with the report they came from, and shown again until **Refresh** brings new data. ## A cluster, opened | Block | What it shows | | --- | --- | | Words | The terms that mark the cluster, as chips. | | Pages that get these clicks | Each page with its clicks and its number of queries in the cluster. | | Queries | Each query with its clicks, impressions and position. Up to 50, with **Showing 50 of N.** | | Split between pages | Shown on the row when two or more pages each take at least 15% of the cluster's clicks. | A split cluster is worth reading beside [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization), which works query by query, and the crawl's [Topics](https://www.getlooploop.com/gllmolfar/docs/topics), which group the pages themselves. ## States and messages - **Working** — the app says: "Clustering…" On the button while the model works. - **Too few** — the app says: "Too few queries to cluster." The report has too few queries to group. ## Related pages - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Cannibalization](https://www.getlooploop.com/gllmolfar/docs/gsc-cannibalization.md): GSC › Cannibalization: queries for which Google showed two or more of your pages, with each page's clicks, share of impressions and position. - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. - [Search Console: Visualisation](https://www.getlooploop.com/gllmolfar/docs/gsc-visualisation.md): GSC › Visualisation: the period's queries drawn: words as a treemap, spread by position, CTR and impressions, opportunities, click concentration, brand. --- # Search Console: Visualisation > The period's queries drawn instead of listed: the words they are made of, how they spread over position, CTR and impressions, position against CTR with the queries one step from more clicks, how concentrated clicks are, query length, and brand and questions against the rest. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/gsc-visualisation - **Where:** GSC › Visualisation - **Mode:** Simple and Pro - **Built from:** The queries the saved report already holds; nothing is fetched - **Changes:** With the period, and on Refresh ## On the screen The charts follow the period buttons in the shared head (see [Performance](https://www.getlooploop.com/gllmolfar/docs/search-console)). The first is the treemap of words; the others follow below it. ![The Visualisation report of the GSC section in GLLmolfar for getlooploop.com, with the treemap of words in queries](https://www.getlooploop.com/images/gllmolfar/docs/gsc-viz.webp) Numbered parts of the screenshot: 1. **Period.** The dates against the period before, and when the report was read. 2. **Basis.** How many queries Search Console lists for the period and their share of its clicks. Google leaves out rare queries. 3. **Words in queries.** Every query split into words; a word's tile adds up the queries that contain it. 4. **Clicks, Impressions, Queries.** What the tiles are sized and shaded by. 5. **A word.** Darker for more. Click it to see **Queries with “{word}”**. 6. **Placeholder names.** Competitor names in queries are replaced in this picture. ## The charts | Chart | What it shows | | --- | --- | | Words in queries | A treemap of the words in the queries, without words that carry no topic. Sized by clicks, impressions or the number of queries. | | Queries by position | Queries in five bands, 1–3 to 51+. Number of queries, log scale. | | Queries by CTR | Queries by click-through rate, 0–1% to 10%+. Number of queries, log scale. | | Queries by impressions | Queries by impressions, 0–10 to 100K+. Number of queries, log scale. | | Position against CTR | Each dot is a query, sized by impressions. The shaded band is positions 4–20, where one step up gains the most. Opportunities are drawn in their own colour; click a dot to open the query. | | Opportunities | Queries on positions 4–20 with more impressions than most and a CTR below this site's own top-3 rate. Columns: query, position, CTR and **+ clicks**, the clicks it would add at that rate. Up to 10. | | How concentrated clicks are | Queries sorted by clicks; the line is the running share of all their clicks, the dashed diagonal an even spread. With the share of clicks from the top 10% of queries, and how many queries bring 80% of clicks. | | Query length | Queries by the number of words in them; under each bar, its share of clicks. | | Brand, questions and the rest | Clicks, impressions and queries split into **Brand** (the query contains the property's name), **Questions** (it starts with a question word, the way people ask an assistant) and **Other**. | The same five position bands are listed in [Positions](https://www.getlooploop.com/gllmolfar/docs/gsc-positions). Questions people search are also gathered in [AI](https://www.getlooploop.com/gllmolfar/docs/ai). ## States and messages - **No queries** — the app says: "No queries in this period." Nothing to draw. - **No benchmark** — the app says: "Too few queries on positions 1–3 to know this site's top-3 CTR, so there is nothing to measure the others against." In place of the opportunities. - **No opportunity** — the app says: "No query fits: nothing on positions 4–20 with many impressions and a CTR below the top-3 rate." In place of the opportunities. - **No clicks** — the app says: "No clicks in this period." In place of the concentration chart. - **Short name** — the app says: "The property's name is too short to tell brand queries apart. Questions: the query starts with a question word." Brand is left out of the split. ## Related pages - [Search Console: Performance](https://www.getlooploop.com/gllmolfar/docs/search-console.md): GSC › Performance: Google clicks, impressions, CTR and position read-only, kept on this computer and shown with their date when you are offline or signed out. - [Search Console: Positions](https://www.getlooploop.com/gllmolfar/docs/gsc-positions.md): GSC › Positions: the period's queries in five bands of average position, and the queries on positions 4 to 20 where a small gain moves clicks most. - [Search Console: Query clusters](https://www.getlooploop.com/gllmolfar/docs/gsc-query-clusters.md): GSC › Query clusters: up to 1000 Search Console queries grouped by meaning on this computer, with each cluster's clicks, position, pages and queries. - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. --- # Free tools > The five free GetLoopLoop tools, inside the app. Each one is the tool alone from getlooploop.com — its field, its button and its reading — with a line above it about the app's allowance and a button to open the full page in the browser. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/free-tools - **Where:** Free tools in the rail - **Mode:** Simple and Pro - **Built from:** The tools on getlooploop.com, not the open crawl - **Changes:** Each time a tool is run ## The section **Free tools** on the strip opens a panel with the five tools. Pressing one loads it in the view, with the app's wait bar (**Opening the tool**) until the page is in. The tool speaks the interface language when the site has it — English, Ukrainian or Spanish — and English otherwise. | Tool | What it reads | | --- | --- | | Vectorscope | Paste a public URL to see its readable passages, semantic groups and the content that falls outside the retrieval cap. | | Voicescope | Start with one keyword and explore the questions a person may ask an AI assistant across real-life contexts. | | Brandscope | Ask three AI assistants the buyer questions around a brand, then compare their answers with the brand's own page. | | Reviewscope | Ask three AI assistants what reviews say about a brand — first without naming a platform, then naming each one — and see which ratings they state without a source. | | PatternScope | Paste a page to see the stock phrases, repeated sentence shapes and unchecked claims a reader reacts to. Not an AI detector. | ## The line above each tool | Part | What it says or does | | --- | --- | | Name | The tool open now. | | Allowance note | Runs made in the app are counted on GetLoopLoop's server, on a counter of the app's own, apart from the website's. Signed out: **Without signing in the app has the website's allowance on a counter of its own; signing in counts by your account and its plan.** Signed in: **Runs here count against your account's allowance in the app, apart from the website's; the tool shows what is left.** | | Sign in with Google | Shown while signed out. After signing in, runs count by the account and its plan. | | Open in the browser | Opens the tool's full page on getlooploop.com, with its explanations, in the browser. | > **The pictures.** The lines where a tool states what is left of its allowance are hidden in these pictures. Each tool shows its own while you use it; nothing in the app can raise it. > **Examples are not readings.** Before a run, Voicescope, Brandscope, Reviewscope and PatternScope show an **Illustrative example**: a fictional brand, sample answers or a sample text, so you can see what a reading looks like. Nothing in it was measured. Your own reading replaces it once you run the tool. ## Vectorscope ![Vectorscope inside GLLmolfar, before a page is read](https://www.getlooploop.com/images/gllmolfar/docs/tools-vectorscope.webp) Numbered parts of the screenshot: 1. **Head.** **Vectorscope** and the allowance note. 2. **Sign in, browser.** **Sign in with Google** and **Open in the browser**. 3. **The page.** A public URL, and **Read the page**. **Try example** fills in an example address (replaced in this picture). 4. **Result.** How a retriever reads the page, step by step: **1. Fetched**, **2. Extracted**, **3. Chunked**, **4. Embedded**, **5. Grouped**, with passages read, slots used of the passage cap, unique text, and characters past the cap. 5. **Before a reading.** A drawing of one page cut on its headings: the passages inside the cap are read, the ones past it are dropped. ## Voicescope ![Voicescope inside GLLmolfar with its illustrative example](https://www.getlooploop.com/images/gllmolfar/docs/tools-voicescope.webp) Numbered parts of the screenshot: 1. **Keyword.** **a keyword you already rank for**, and **Expand the key**. **Try:** offers a few keywords to start from. 2. **Illustrative example.** **Generated question hypotheses, not observed searches or demand.** Filters by context label and keyword, and **Map** or **Questions**. 3. **Map.** The keyword in the middle, the jobs a person is doing around it, and the questions each job asks. 4. **Selected job.** The questions of the job picked on the map, with the contexts each one crosses. 5. **Why this question?.** The job, the contexts crossed, how the question stands to the keyword, and its place in the job. ## Brandscope ![Brandscope inside GLLmolfar with its illustrative example](https://www.getlooploop.com/images/gllmolfar/docs/tools-brandscope.webp) Numbered parts of the screenshot: 1. **Brand.** **Brand or website**, **Add a category (optional)**, **Run the scan**, and **Show an example**. 2. **Counts.** **Your page**, **ChatGPT**, **Claude**, **Gemini** and **Knows you**: dashes until a scan has run. 3. **Illustrative example.** **Fictional brand, sample answers — nothing here was measured. Run a scan to read your own.** 4. **Questions.** **What is it?**, **Pricing**, **Integrations**, **Weaknesses**: the buyer question whose answers are shown, beside what the brand's own page says. 5. **Why this verdict?.** The verdict on one answer, here **Contradicted**, with the evidence from the brand's page and from the assistant's answer side by side. ## Reviewscope ![Reviewscope inside GLLmolfar with its illustrative example](https://www.getlooploop.com/images/gllmolfar/docs/tools-reviewscope.webp) Numbered parts of the screenshot: 1. **Brand.** **Brand or website**, **Add a category (optional)**, **Run the probe**, and **Show an example**. 2. **Counts.** **Own page**, **ChatGPT**, **Claude**, **Gemini** and **Reading**: dashes until a probe has run. 3. **Illustrative example.** **Fictional brand Acme and sample assistant claims — not a reading.** 4. **Three readings.** **Unprompted**: which platforms they mention. **When asked**: what they claim. **Invented**: ratings for platforms that do not exist. **Show numerical claims only** narrows the claims. 5. **Platform by platform.** Each platform against each assistant: the rating it claims, unchecked, or **Declined to guess**. The platforms' names are replaced in this picture. ## PatternScope ![PatternScope inside GLLmolfar with its illustrative example](https://www.getlooploop.com/images/gllmolfar/docs/tools-patternscope.webp) Numbered parts of the screenshot: 1. **Your text.** **Paste a page, a post or a draft.** The count of words used sits in the corner of the field. 2. **Illustrative example.** **Sample text, not your reading**, with the number of findings and **Density, not a quality grade.** 3. **Marked text.** Each finding highlighted in the text, in its kind's colour. 4. **Kinds.** **All findings**, **Stock phrasing**, **Uniform openings** and **Unchecked claims**, each with its count; **Highlight** turns the marks off and on. 5. **Why marked?.** One finding explained, with its rule, **Dismiss finding** and **Next finding**. ## States and messages - **Loading** — the app says: "Opening the tool" The wait bar, until the tool's page is in. - **Signed out** — the app says: "Without signing in the app has the website's allowance on a counter of its own; signing in counts by your account and its plan." The app's own counter, apart from the website's. **Sign in with Google** sits beside it. - **Signed in** — the app says: "Runs here count against your account's allowance in the app, apart from the website's; the tool shows what is left." Runs count by the account and its plan. ## Related pages - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. --- # Settings › General > How the app looks and behaves: its language, whether it explains technical terms, its mode, what a new crawl starts with, whether it checks for updates and sends anonymous statistics, what the close button does, and whether AI agents may use it. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/settings-general - **Where:** Settings › General - **Opens:** The gear at the top right - **Stored:** In the settings file on this computer - **Changes:** When you press Save ## On the screen ![Settings › General in GLLmolfar](https://www.getlooploop.com/images/gllmolfar/docs/settings-general.webp) Numbered parts of the screenshot: 1. **Interface language.** English, Українська, Español, Français or Polski. 2. **Explain technical terms.** **Explain technical terms (nosnippet, X-Robots-Tag…) on hover**: the dotted underlines and their one-line explanations. 3. **Max URLs and interface mode.** The number the toolbar starts with, and **Simple mode** or **Pro mode**. 4. **Updates, statistics, tray.** Three switches: a new version at start, anonymous usage statistics, and the close button. 5. **AI agents (MCP).** Lets AI agents on this computer use the app, with **How to connect an agent…** and **Agent skills…**. 6. **Sound.** **Sounds when a crawl and its embeddings finish**. 7. **Default mode.** **Spider + sitemap**, **Spider only** or **Sitemap only**. ## Fields | Field | What it does | Options | Default | | --- | --- | --- | --- | | Interface language | The language of the whole app. | English, Українська, Español, Français, Polski | English | | Explain technical terms (nosnippet, X-Robots-Tag…) on hover | Terms such as nosnippet, X-Robots-Tag, hreflang, llms.txt or LCP get a dotted underline wherever they appear. Hovering or focusing one shows one sentence in the interface language and **More in the GetLoopLoop glossary**. Off: the underlines go. | On or off | On | | Default max URLs | How many URLs a crawl reads, as the **Max URLs** box in the toolbar starts. | A number | 500 | | Interface mode | Simple shows what to do and why; Pro shows every table and filter. | Simple mode, Pro mode | Simple mode | | Check for a new version at start | Asks getlooploop.com for the latest version when the app starts. | On or off | On | | Send anonymous usage statistics | **A random install ID, app version, Windows version, language and how long the app is open. Never URLs or crawl data.** Messages from GetLoopLoop arrive only in reply, so switching this off stops those too. | On or off | On | | The close button keeps GLLmolfar running in the tray | On: closing the window hides it to the tray, and planned crawls can still run. Off: closing the window quits the app. | On or off | On | | Let AI agents use GLLmolfar on this computer (MCP) | AI agents may drive the app through its MCP server on this computer. See [MCP tools](https://www.getlooploop.com/gllmolfar/docs/mcp-tools). | On or off | Off | | How to connect an agent… | Shows how to connect an AI agent to the app. | Button | — | | Agent skills… | Opens the agent skills. | Button | — | | Sounds when a crawl and its embeddings finish | Plays a soft sound at the end of a crawl. | On or off | On | | Default mode | How a crawl finds pages, as the **Mode** box in the toolbar starts. | Spider + sitemap, Spider only, Sitemap only | Spider + sitemap | > **The diagnostic log in a bug report.** Usage statistics are not the only thing that can leave the computer. A **Bug** in **Feedback** can carry the diagnostic log: **Attach the diagnostic log** is ticked by default, with **Events from the last** 30 minutes, an hour or 2 hours, and **Show the log** to read it first. **Kept only on this computer, for two hours. Counts and errors only: no addresses, domains, page text, IPs, emails or paths in your user folder. It leaves the computer only with this report.** > **Locked switches.** The MCP switch can be locked. A lock means it is not open on this account; pressing it says what opens it. ## States and messages - **Saved** — the app says: "Settings saved." Save closes the dialog. Cancel, Escape or a click outside it closes it without saving. - **Next crawl** — the app says: "Crawl and embedding settings apply to the next crawl." A crawl that is running keeps the settings it started with. ## Related pages - [Settings › Crawling](https://www.getlooploop.com/gllmolfar/docs/settings-crawling.md): Settings › Crawling: request speed, connections, the user agent, the timeout, robots.txt, a pause at a CAPTCHA, JavaScript rendering and service pages. - [Settings › Crawl scope](https://www.getlooploop.com/gllmolfar/docs/settings-scope.md): Settings › Crawl scope: which URLs a crawl may visit — include and exclude patterns, depth, URLs with parameters, subdomains, nofollow and external links. - [Settings › Planned crawls](https://www.getlooploop.com/gllmolfar/docs/settings-planned.md): Settings › Planned crawls: what happens to a run the computer was off for, starting with the system in the tray, and notifications for planned crawls. - [MCP tools for agents](https://www.getlooploop.com/gllmolfar/docs/mcp-tools.md): The MCP server inside GLLmolfar: how to connect an AI agent, and every tool it can call to crawl, read fixes, audit pages, search by meaning and save reports. --- # Settings › Crawling > How the crawler asks a site for its pages: how fast, as which user agent, how long it waits, which robots.txt it obeys, what it does at a CAPTCHA, and when it renders JavaScript. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/settings-crawling - **Where:** Settings › Crawling - **Opens:** The gear at the top right - **Stored:** In the settings file on this computer - **Applies:** To the next crawl ## On the screen ![Settings › Crawling in GLLmolfar](https://www.getlooploop.com/images/gllmolfar/docs/settings-crawling.webp) Numbered parts of the screenshot: 1. **Speed.** **Requests per second** and **Parallel connections**. 2. **User-Agent.** Who the crawler says it is: GLLmolfar, a search engine bot, Chrome, an AI crawler or your own. 3. **Respect robots.txt.** Obey the site's robots.txt. 4. **At a CAPTCHA.** **At a CAPTCHA, pause and open it for me to pass**: the crawl waits while you pass the site's check, then goes on. 5. **Custom robots.txt.** A robots.txt to use instead of the site's, and **Test URL** to try an address against it. 6. **JavaScript rendering.** Off, pages with little text, or all pages. 7. **Service pages.** **Count service pages in content checks**. ## Fields | Field | What it does | Options | Default | | --- | --- | --- | --- | | Requests per second | The most requests the crawler sends each second. | 0.2 to 20 | 5 | | Parallel connections | How many requests run at once. | 1 to 16 | 4 | | User-Agent | The name the crawler gives the site. A bot preset obeys that bot's own group in robots.txt. | GLLmolfar; Googlebot Smartphone; Bingbot; Chrome (Windows); under AI: GPTBot (OpenAI), OAI-SearchBot (OpenAI), ChatGPT-User (OpenAI), ClaudeBot (Anthropic), PerplexityBot (Perplexity); Custom… | GLLmolfar | | Custom… (text box) | Shown for **Custom…**: your own user agent. It cannot name another company: Save is refused, the box gets the focus, and the section says **A custom user agent cannot name another company**, with the word it found. | Text | Empty | | Request timeout, s | Seconds to wait for one answer before giving up. | 3 to 120 | 15 | | Respect robots.txt | Skips addresses the site's robots.txt closes. With a bot preset as the user agent, robots.txt is obeyed even when this is off. | On or off | On | | At a CAPTCHA, pause and open it for me to pass | When a page answers with a CAPTCHA or a bot check, in place or by sending every page to a check page, the crawl pauses and the page opens in a window of its own, with the crawl's user agent. Once the site lets you through, the crawl resumes and asks for the page again with the cookies the check left. Off, and in planned crawls: such pages are recorded as bot checks and the crawl goes on. | On or off | On | | Custom robots.txt | **Used instead of the site's robots.txt, to try changes before they go live. Leave empty to use the site's.** | Text | Empty | | Test URL | Checks one address against the custom robots.txt, or against the site's when the box is empty. Without an address it tests the one in the toolbar. | Button | — | | JavaScript rendering | Whether pages are opened in a browser so that text written by JavaScript is seen. | Off · raw HTML only; Auto · render pages with little text; All pages (slower) | Auto · render pages with little text | | Count service pages in content checks | **Off: privacy, terms, login, cart, the HTML sitemap and similar pages are left out of the author, date, thin content, cannibalization and other content checks. Each check says how many it left out.** | On or off | Off | ## States and messages - **Allowed** — the app says: "Allowed · custom robots.txt" **Test URL**: the address may be crawled. The second part says which robots.txt answered: **custom robots.txt**, **the site's robots.txt** or **the site has no robots.txt**. - **Blocked** — the app says: "Blocked · the site's robots.txt" **Test URL**: the robots.txt closes this address. - **Next crawl** — the app says: "Crawl and embedding settings apply to the next crawl." A crawl that is running keeps the settings it started with. ## When a site asks for a CAPTCHA With the switch on, the first check a crawl meets pauses it and opens a window titled **Pass the site's check — GLLmolfar**. A bar in the main window says what happened and offers three buttons. - **Paused** — the app says: "The crawl is paused: the site asks for a CAPTCHA." Under it: **Pass it in the window that opened; the crawl goes on by itself once the site lets you through.** When the app recognises the check, its maker's name comes first. | Button | What it does | | --- | --- | | Show the window | Brings the check's window to the front. | | I passed it, continue | Resumes the crawl when the window has not noticed on its own. | | Skip checks in this crawl | Closes the window; this and later checks are recorded as bot checks until the crawl ends. Closing the check's window does the same. | ## Related pages - [Settings › Crawl scope](https://www.getlooploop.com/gllmolfar/docs/settings-scope.md): Settings › Crawl scope: which URLs a crawl may visit — include and exclude patterns, depth, URLs with parameters, subdomains, nofollow and external links. - [Settings › Requests](https://www.getlooploop.com/gllmolfar/docs/settings-requests.md): Settings › Requests: extra HTTP headers sent with every page request, and a user name and password for a staging site behind Basic authentication. - [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general.md): Settings › General: the interface language, term explanations, the interface mode, the default crawl size and mode, updates, statistics, the tray, MCP, sound. - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. --- # Settings › Crawl scope > Which addresses a crawl may visit. Patterns keep or drop URLs, a depth stops it some clicks from the start, and three switches decide about subdomains, nofollow links and external links. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/settings-scope - **Where:** Settings › Crawl scope - **Opens:** The gear at the top right - **Stored:** In the settings file on this computer - **Applies:** To the next crawl ## On the screen ![Settings › Crawl scope in GLLmolfar](https://www.getlooploop.com/images/gllmolfar/docs/settings-scope.webp) Numbered parts of the screenshot: 1. **Include.** Only URLs that match one of these lines are crawled. 2. **Exclude.** URLs that match any of these lines are skipped. 3. **Max crawl depth.** Clicks from the start URL; 0 is no limit. 4. **URLs with parameters.** **Crawl as found**, **Remove the parameters** or **Skip them**. 5. **Subdomains.** **Crawl subdomains as part of the site**. 6. **Nofollow.** **Follow internal nofollow links**. 7. **External links.** **Check external links**. ## Fields | Field | What it does | Options | Default | | --- | --- | --- | --- | | Include (regex, one per line) | When any line is given, a URL must match one of them to be crawled. Example: `/blog/`. | Regular expressions | Empty | | Exclude (regex, one per line) | A URL that matches any line is skipped. Example: `\?sort=\|/tag/`. | Regular expressions | Empty | | Max crawl depth (0 = no limit) | How many clicks from the start URL the crawl follows. | 0 to 100 | 0 | | URLs with parameters | What to do with an address that has a query string (`?…`). | Crawl as found; Remove the parameters; Skip them | Crawl as found | | Crawl subdomains as part of the site | Treats subdomains of the start host, such as blog.example.com, as the same site. | On or off | Off | | Follow internal nofollow links | Follows internal links marked `rel="nofollow"`. | On or off | On | | Check external links | Requests each external link once and records its status. | On or off | On | > **What the rules leave alone.** The start URL and URLs you add yourself are never dropped by include, exclude, parameters or depth. Parameters are handled first, so with **Remove the parameters** the patterns see the address without its query string. ## States and messages - **Bad pattern** — the app says: "This line is not a valid regular expression: {detail}" Save is refused; the message shows under the switches and the Include box gets the focus. - **Next crawl** — the app says: "Crawl and embedding settings apply to the next crawl." A crawl that is running keeps the scope it started with. ## Related pages - [Settings › Crawling](https://www.getlooploop.com/gllmolfar/docs/settings-crawling.md): Settings › Crawling: request speed, connections, the user agent, the timeout, robots.txt, a pause at a CAPTCHA, JavaScript rendering and service pages. - [Settings › Requests](https://www.getlooploop.com/gllmolfar/docs/settings-requests.md): Settings › Requests: extra HTTP headers sent with every page request, and a user name and password for a staging site behind Basic authentication. - [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general.md): Settings › General: the interface language, term explanations, the interface mode, the default crawl size and mode, updates, statistics, the tray, MCP, sound. - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. --- # Settings › Requests > What GLLmolfar adds to each request it sends. Extra headers, such as a language, and a user name and password for a site that asks for Basic authentication. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/settings-requests - **Where:** Settings › Requests - **Mode:** Simple and Pro - **Stored:** In the settings file on this computer - **Changes:** From the next crawl ## On the screen ![Settings › Requests in GLLmolfar: a box for extra HTTP headers and two fields for Basic authentication](https://www.getlooploop.com/images/gllmolfar/docs/settings-requests.webp) Numbered parts of the screenshot: 1. **Sections.** The Settings sections. **Requests** is the fourth. 2. **Extra HTTP headers.** One header per line, as `Name: value`. 3. **Basic authentication.** A user name and a password for a site behind a password. 4. **Where it is kept.** **For staging sites behind a password. Kept on this computer only, in the settings file.** 5. **When it applies.** **Crawl and embedding settings apply to the next crawl.** 6. **Cancel and Save.** Nothing changes until **Save**. ## Fields | Field | What it does | Default | | --- | --- | --- | | Extra HTTP headers (Name: value) | Each line becomes one header on every request of the crawl. The name is the text before the first colon, the value the text after it. A line without a colon is ignored. Example: `Accept-Language: uk`. | Empty | | Basic authentication: user | With a user name set, every request carries an `Authorization: Basic` header made from the user name and the password. | Empty | | Basic authentication: password | The password for that header. The field hides what you type. | Empty | > **Where the password lives.** The user name and the password stay in the settings file on this computer. Leave the user name empty to send no `Authorization` header at all. ## States and messages - **Saved** — the app says: "Settings saved." The new headers and credentials go out with the next crawl. ## Related pages - [Settings › Crawling](https://www.getlooploop.com/gllmolfar/docs/settings-crawling.md): Settings › Crawling: request speed, connections, the user agent, the timeout, robots.txt, a pause at a CAPTCHA, JavaScript rendering and service pages. - [Settings › Crawl scope](https://www.getlooploop.com/gllmolfar/docs/settings-scope.md): Settings › Crawl scope: which URLs a crawl may visit — include and exclude patterns, depth, URLs with parameters, subdomains, nofollow and external links. - [Settings › Storage](https://www.getlooploop.com/gllmolfar/docs/storage.md): Settings › Storage: how much disk the saved crawls, the embedding model and downloaded updates take, per project, and the actions that free it. - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses. --- # Settings › Embeddings > Embeddings are what topics, cannibalization, semantic search and the topic map are built from. This section sets whether a new crawl makes them, how much of each page the model reads, and where the model runs. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/settings-embeddings - **Where:** Settings › Embeddings - **Model:** One model on this computer, the same in every mode - **Built from:** The text of each crawled page - **Changes:** From the next crawl ## On the screen ![Settings › Embeddings in GLLmolfar: four mode cards under a term explanation, the test button, and the model's text, thread and device settings](https://www.getlooploop.com/images/gllmolfar/docs/settings-embeddings.webp) Numbered parts of the screenshot: 1. **A term explained.** Hovering the dotted **Embeddings** heading: one sentence and **More in the GetLoopLoop glossary**. Under it, **Embeddings on for new crawls**. 2. **Modes.** **Standard**, **Extended**, **Long** and **Longest**: how much text per section and how many sections. Each card shows its result in our test and, once tested, its result on this computer. 3. **Test on this computer.** Measures every mode on the open crawl's longest pages. Nothing leaves the computer. 4. **Locked modes.** **Modes beyond standard are for signed-in accounts (free).** **Sign in with Google** stands beside it, and the cards carry a lock. 5. **The model's settings.** **Text per page** and **Long pages**, the same choice as the cards; then **CPU threads for the model** and **Where the model runs**. 6. **Test and tune CPU and GPU.** Measures the CPU and the GPU and sets threads and batch sizes for this computer. ## The four modes Every mode runs the same model, so changing mode downloads nothing. A mode is a size of text per section and a number of sections per page, printed on the card as `tokens × sections`. **Our test** is a fixed set of pages: a sentence from the second half of each page had to find its own page, and the figure is how often that page was in the first five. | Card | Size | What the card says | Our test: right page in the top 5 | | --- | --- | --- | --- | | Standard | 256 × 4 | The start of each of four sections. Fastest. | 71% | | Extended | 512 × 4 | Twice the text of each section. The best match in our test. | 80% | | Long | 1024 × 4 | Four times the text. Slower, no better in our test. | 79% | | Longest | 2048 × 2 | Two sections read almost whole. Slowest. | 77% | After **Test on this computer**, each card shows how many pages per second it embedded here, how many times slower than Standard it was, and how often it **found its page** among the open crawl's pages. Three marks can appear: **Most accurate here**, **Fastest here** and **Found fewer here**. The result is kept until the next test, with the site, the number of pages and the date. ## Fields | Field | Options | What it does | | --- | --- | --- | | Embeddings on for new crawls | On or off. On by default. | Whether the toolbar's **Embeddings** box starts ticked. Without embeddings a crawl has no topics, cannibalization or semantic search. | | Text per page | 256 tokens (faster), 512 tokens (more precise), 1024 tokens, 2048 tokens (slowest) | How much text of each section the model reads. | | Long pages | Start of the page only (fastest), Up to 2 sections, Up to 4 sections (recommended), Up to 8 sections (slowest) | How many sections of a long page are read. | | CPU threads for the model | From 1 to the number of cores. Half the cores by default. | How many processor threads the model uses. | | Where the model runs | Balanced: the faster of CPU and GPU (recommended), CPU only, GPU (DirectML) | Balanced uses the CPU until a test shows the GPU is clearly faster. The GPU choice and the test exist on Windows only; elsewhere the model runs on the CPU. | | Test and tune CPU and GPU | A button | Tries thread counts and batch sizes on the CPU, then the GPU, in about two minutes, and keeps the best for this computer. | ## States and messages - **Not tested** — the app says: "Not tested on this computer yet" A card before the first test. Only our test's figure is shown. - **Testing** — the app says: "Mode 2 of 4: 512 × 4…" The mode test is running, one mode after another. - **Busy** — the app says: "Wait until the crawl and its embeddings finish, then test." A test cannot run beside a crawl. - **No pages** — the app says: "Open a crawl with saved pages first: the test uses its longest pages." The mode test needs an open crawl. - **Failed** — the app says: "The test stopped: …" The reason follows the colon. Settings are unchanged. - **Device not tested** — the app says: "Not tested yet. Balanced mode uses the CPU until a test shows the GPU is clearly faster." Before **Test and tune CPU and GPU** has run. - **Device testing** — the app says: "Testing thread counts and batch sizes on the CPU, then the GPU: about two minutes…" The device test is running. - **Stays on CPU** — the app says: "Balanced mode stays on the CPU: the GPU is not 1.5× faster, and a built-in GPU takes its memory from the same RAM." The test's verdict, with the measured rates before it. - **Uses GPU** — the app says: "Balanced mode uses the GPU: it is worth its memory here." The test's verdict, with the measured rates before it. - **GPU failed** — the app says: "The GPU could not run the model here; CPU … texts/s. Balanced mode uses the CPU." The model runs on the CPU. ## Related pages - [Topics](https://www.getlooploop.com/gllmolfar/docs/topics.md): The Topics tab: the site's pages grouped by meaning, with each topic's size, key terms, cohesion, internal linking, cannibal pairs and central page. - [Cannibalization](https://www.getlooploop.com/gllmolfar/docs/cannibalization.md): The Cannibalization tab: groups of pages that say almost the same thing, the page to keep, the pages side by side and their text passage by passage. - [Semantic search](https://www.getlooploop.com/gllmolfar/docs/semantic-search.md): The Semantic search tab: type a question and get the crawled pages closest to it in meaning, with the sections of each page that match best. - [Topic map](https://www.getlooploop.com/gllmolfar/docs/topic-map.md): The Topic map tab: every embedded page as a dot, placed by meaning and coloured by topic, status, indexability or depth, with a card for each page. - [Settings › Storage](https://www.getlooploop.com/gllmolfar/docs/storage.md): Settings › Storage: how much disk the saved crawls, the embedding model and downloaded updates take, per project, and the actions that free it. --- # Settings › AI models > GLLmolfar can use your own language models for its written analyses. Add a connection, check it, and choose which model does which job. The section is marked **alpha**. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/settings-ai-models - **Where:** Settings › AI models - **Stage:** Alpha - **Keys:** Encrypted on this computer, never shown again - **Changes:** Saved at once, apart from the Settings form ## On the screen ![Settings › AI models in GLLmolfar with no connection yet: the intro and a button for each kind of provider](https://www.getlooploop.com/images/gllmolfar/docs/settings-ai-models.webp) Numbered parts of the screenshot: 1. **Alpha and the intro.** **Connect your own models for analyses.** Where the page text goes, and how keys are kept. 2. **Add a connection.** One button per supported provider, each with its own mark, and **Other OpenAI-compatible** for any other server. 3. **Connections.** **No models connected yet. Add one above.** until the first one is added; then a card per connection. 4. **When it applies.** **Crawl and embedding settings apply to the next crawl.** The connections themselves are saved as you change them. 5. **Cancel and Save.** For the rest of the Settings form. Connections and routes do not wait for **Save**. ## Where the text goes **A model on this computer or this network keeps the page text here; a cloud provider receives the text an analysis sends it. Keys are encrypted on this computer and never shown again.** Each card says which of the two it is: **On this computer or network**, or **Cloud:** and the host. ## A connection's card Pressing a provider's button adds a card for it, with the provider's usual address filled in. Up to twelve connections can be kept. | Part | What it shows or does | | --- | --- | | Header | The provider's mark and name, **On this computer or network** or **Cloud: host**, and **Remove**. Removing a connection also clears every job that used it. | | Address (OpenAI-compatible, ends in /v1) | Where the server answers. It must start with `http://` or `https://`. | | API key | Shown as **API key** where the provider needs one, **API key (if the server asks for one)** where it may not, and not shown where it is never used. A saved key reads **Saved — type to replace**. | | State | **Not checked yet**, **Connected · models: N · N ms** with the date of the check, or the reason it failed. | | Model list | After a successful check: the models the server offers. The model picked here is the one **Try a reply** asks. | | Check | Saves the address and key as typed, then asks the server for its models. Enter in a field does the same. | | Try a reply | After a successful check: asks the model for one short sentence and shows the reply with its time. A cloud card adds **A few tokens, billed to your key.** | | What the model was thinking | When the model sent its reasoning with the reply, it folds under the reply. | ## Which model does what Once a connection exists, a list of jobs appears under the cards. Each job takes one model from a checked connection, or **— not set —**. The note above it says: **Alpha: the choice is saved now; the analyses start using it in the next builds. Only checked connections are offered.** | Job | What the app says it does | | --- | --- | | Topic names | Names each topic from its pages' titles and terms. | | Explaining fixes | Why a problem matters and how to fix it, in plain words. | | Page audit | A written summary of a single page's audit. | | Reverse prompts | Judges a page's blocks and drafts their prompts. | | Agent (test section) | Plans the work, ranks the tasks and checks its own findings. | | Writing | Rewrites pages and writes articles in the agent's test section. | | Fact checking | Checks every claim of a draft against its source, and judges pages flagged as risky. | | Social posts | Posts, threads and articles for social networks. | ## States and messages - **Connected** — the app says: "Connected · models: 12 · 140 ms" The server answered and listed its models. The connection can now be chosen for a job. - **Needs a key** — the app says: "Reachable, but it needs your API key before it can answer." Type the key and press **Check** again. - **Key refused** — the app says: "The key was refused." The server rejected the key. - **Timeout** — the app says: "No answer within 10 seconds." The server did not answer the check in time. - **Nothing there** — the app says: "Nothing is listening at this address. Is the server running?" Usually a local server that is not started. - **No such host** — the app says: "This address does not resolve." The host name in the address is wrong. - **Other** — the app says: "Could not connect: …" Any other failure, with its detail. - **Only thinking** — the app says: "The model only thought and gave no answer within the limit. Thinking models need more room: it works in the agent, which allows long answers." **Try a reply** allows a short answer only. - **Cut** — the app says: "The answer was cut by the length limit before any text came out." From **Try a reply**. - **Empty** — the app says: "The server answered with empty text. Check that this model writes text (not an embedding model) and that it is loaded." From **Try a reply**. - **No model** — the app says: "No checked model: connect one in Settings › AI models." Shown by an analysis that needs a model when none is checked. ## Related pages - [AI](https://www.getlooploop.com/gllmolfar/docs/ai.md): The AI screen: an AI readiness score, which AI bots may get in and what they get, what blocks quoting, trust signals, llms.txt, Markdown for agents. - [Agent · test](https://www.getlooploop.com/gllmolfar/docs/agent-test.md): The Agent · test screen: your own models plan tasks from the fix list, rewrite pages, write social posts and review risky pages; GLLmolfar measures the results. - [Reverse prompts](https://www.getlooploop.com/gllmolfar/docs/reverse-prompts.md): Site › Reverse prompts: GLLmolfar reads a page into blocks, your AI agent describes what each does, and every block gets a reusable prompt and an example. - [MCP tools for agents](https://www.getlooploop.com/gllmolfar/docs/mcp-tools.md): The MCP server inside GLLmolfar: how to connect an AI agent, and every tool it can call to crawl, read fixes, audit pages, search by meaning and save reports. - [Settings › Embeddings](https://www.getlooploop.com/gllmolfar/docs/settings-embeddings.md): Settings › Embeddings: whether new crawls build embeddings, how much of each page the model reads, and whether it runs on the CPU or the GPU. --- # Settings › Planned crawls > Planned crawls run on this computer while GLLmolfar is open or in the tray. This section decides what happens to a run that came due while the computer or the app was off, and whether the app tells you about each run. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/settings-planned - **Where:** Settings › Planned crawls - **Mode:** Simple and Pro - **Applies to:** Every schedule in Projects › Schedule - **Changes:** When you press Save ## On the screen ![Settings › Planned crawls in GLLmolfar: the missed-run choice and two switches](https://www.getlooploop.com/images/gllmolfar/docs/settings-planned.webp) Numbered parts of the screenshot: 1. **A run the computer was off for.** Run it when the computer is back, skip it, or skip it and remind you. 2. **Start with the system.** **Start with the system, in the tray**, so planned crawls keep running after a restart. 3. **Notifications.** **Notify when a planned crawl starts, finishes or is missed.** 4. **When it applies.** **Crawl and embedding settings apply to the next crawl.** 5. **Cancel and Save.** Nothing changes until **Save**. ## Fields | Field | Options | What it does | | --- | --- | --- | | A run the computer was off for | Run it as soon as the computer is back (default), Skip it, Skip it and remind me | A run that came due while the computer was off or asleep, or the app was closed, is noticed at the next check. **Run it** puts it in the queue. **Skip it** marks it **Skipped** in the calendar. **Skip it and remind me** marks it **Missed** and sends a notice. | | Start with the system, in the tray | On or off. Off by default. | GLLmolfar starts when you sign in to the computer, hidden in the tray. | | Notify when a planned crawl starts, finishes or is missed | On or off. On by default. | System notifications for each planned run: started, waiting, done with its counts, missed, or did not start. | > **The close button matters too.** Planned crawls cannot run once the app is closed. **The close button keeps GLLmolfar running in the tray** is in [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general); Projects › Schedule warns when it is off. ## Notifications With notifications on, each comes under the title **GLLmolfar · planned crawl**. - **Started** — the app says: "example.com: planned crawl started, up to N pages" The run began. - **Waiting** — the app says: "example.com: the planned crawl waits for the crawl in progress to finish" A crawl was already running. A planned run never interrupts it. - **Done** — the app says: "example.com: done — N pages, N errors, N non-indexable" Each figure carries its change against the crawl before, in brackets, when it changed. - **Missed** — the app says: "example.com: the run of … was missed — the computer or the app was off" Only with **Skip it and remind me**. - **Failed** — the app says: "example.com: the planned crawl did not start (code)" The code says why. ## Related pages - [Schedule](https://www.getlooploop.com/gllmolfar/docs/schedule.md): The Schedule tab: a month calendar of planned and past crawls, the list of schedules, and the dialog that plans a crawl once or on repeat. - [Settings › General](https://www.getlooploop.com/gllmolfar/docs/settings-general.md): Settings › General: the interface language, term explanations, the interface mode, the default crawl size and mode, updates, statistics, the tray, MCP, sound. - [History](https://www.getlooploop.com/gllmolfar/docs/history.md): The History tab: every saved crawl of a site, a trend of one number across them, and a comparison of two crawls page by page and check by check. - [Projects](https://www.getlooploop.com/gllmolfar/docs/projects.md): The Projects tab: one card per site crawled on this computer, with its health, how the last crawl changed, its scans over time, topics and domain. --- # Settings › Storage > How much space GLLmolfar takes on this disk, and per project. Each project can drop its older crawls or their saved HTML, and the database can be compacted to give freed space back to the disk. Describes GLLmolfar 0.9.22, checked 2026-11-01. Canonical page: https://www.getlooploop.com/gllmolfar/docs/storage - **Where:** Settings › Storage - **Space on this disk:** Since 0.9.16 - **Built from:** The crawl database on this computer - **Changes:** Each time Settings opens ## On the screen ![Settings › Storage in GLLmolfar with the getlooploop.com project](https://www.getlooploop.com/images/gllmolfar/docs/storage.webp) Numbered parts of the screenshot: 1. **Data folder.** Where the crawls and settings live, and **Open folder**. 2. **HTML and export.** **Keep each page's HTML for the Source tab**, the export format (Excel, CSV or Google Sheets) and whether to ask at every export. 3. **Saved crawls.** The size of the crawl database and **Compact**, with the space freed inside it when there is any. 4. **Model, updates, free space.** The embedding model, downloaded installers with **Remove** when there are any, and what the disk has left. 5. **Projects.** Every saved site, largest first: crawls, the latest, the size and how much of it is HTML, with the actions for that project. 6. **Delete saved crawls older than.** Never, 30 days, 90 days or 1 year. ## What is counted | Row | What it shows | | --- | --- | | Data folder | The folder that holds the crawl database, the model and the settings. **Open folder** shows it in the file manager. | | Keep each page's HTML for the Source tab | On: every crawl saves each page's source, so the Source tab can show it. It is the largest part of a crawl. | | Export format | Excel (.xlsx), CSV or Google Sheets, with **Ask Excel, CSV or Google Sheets at every export**. Google Sheets makes a new spreadsheet in your Drive; signed out, the section says **Google Sheets: sign in to GLLmolfar first.** | | Delete saved crawls older than | Never, 30 days, 90 days or 1 year. | | Saved crawls (database) | The size of the crawl database file. **N freed inside the file; Compact gives it back to the disk** when deletions left free space. After the first compaction: **Deleted data shrinks the file on its own**. | | Embedding model | Downloaded once; needed for topics and cannibalisation. | | Downloaded updates | Installers of versions already installed; safe to remove. The app also removes them itself after an update. | | Free on this disk | Free space on the disk that holds the data folder. | > **Sizes are estimates.** Sizes per project carry ≈. Measuring a large database exactly takes minutes, so the app measures the saved HTML and estimates the rest from row counts and average row sizes. ## Projects and their actions | Column or button | What it shows or does | | --- | --- | | Site | The project's site. | | Crawls | How many crawls of it are saved. | | Latest | When the newest one ran. | | Size | What the project takes, ≈. | | of it HTML | How much of that is saved page HTML. | | Keep latest 3 | Deletes every crawl but the three newest. Asks first, with the number of crawls and the space. | | Old HTML | Deletes the saved HTML of every crawl but the newest. The crawls stay with their pages, links, findings and history. | | Delete | Deletes the project and all its crawls. | ## States and messages - **Counting** — the app says: "Counting links · 12 s" A bar fills step by step (pages, links, page vectors, fragments, external links, link checks, saved HTML), with the step and the seconds. Under it: **On a large database this takes up to a minute. Nothing is changed; you can close Settings meanwhile.** - **Busy** — the app says: "A deletion or compaction is running; the buttons come back when it ends." Progress shows in a card at the bottom of the window. Crawls cannot start during a compaction. - **Failed** — the app says: "Could not read the sizes. Close the settings and open them again." The sizes could not be read this time. Nothing was changed. - **Sheets refused** — the app says: "Google refused access to Sheets. Connect again in Settings › Storage." An export to Google Sheets that Google turned down for want of permission. - **Sheets off** — the app says: "Google Sheets is switched off for GLLmolfar on Google's side. Connecting again will not help; export to Excel or CSV for now." Since 0.9.18 the two refusals are told apart. ## Related pages - [Overview](https://www.getlooploop.com/gllmolfar/docs/overview.md): The Overview tab after a crawl: the health score, six counters, what to fix first, response codes, indexability, site depth, topics and earlier crawls. - [The crawl screen](https://www.getlooploop.com/gllmolfar/docs/crawl-screen.md): What GLLmolfar shows while a crawl runs: four stages, the crawl and fragment cards, robots.txt rules, the run charts, the last pages read and CAPTCHA pauses.