llms-txt-generator
Skill3 starsChanged 6 months agoMITView on GitHub
Generate a spec-compliant llms.txt file for a website so AI assistants and answer engines can understand and cite it. Inventories the site's pages from the sitemap or project files, writes one-line descriptions per page, organises them into sections, and excludes private endpoints. Use when the user says "llms.txt", "make my site readable to AI", "AI search optimization file", or "GEO file".
6 other projects use a version of this.
Adopted by
Most recently taken up by
- imtiazrayhan/agentscamp-librarytheir own version, first seen Nov 16, 2026
- solana-foundation/solana-comtheir own version, first seen Oct 8, 2026
- UXMagicai/seosorted-skillstheir own version, first seen Oct 4, 2026
- GianlucaNaarden/claude-skills-nederlandstheir own version, first seen Aug 10, 2026
- famzila/agent-skills-settheir own version, first seen Mar 25, 2026
- akhilkannur/marketing-agent-blueprintstheir own version, first seen Feb 4, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- solana-foundation/solana-com★ 538same name
- akhilkannur/marketing-agent-blueprints★ 4same name
- famzila/agent-skills-set★ 2same name
- GianlucaNaarden/claude-skills-nederlands★ 1same name
- imtiazrayhan/agentscamp-library★ 1same name
- UXMagicai/seosorted-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository47 other files it gives its agents.
- agents-md-writerSkill
- ai-image-run-pacingSkill
- automation-state-reconcilerSkill
- backup-readiness-auditorSkill
- brand-voice-linterSkill
- case-study-writerSkill
- client-delivery-auditorSkill
- client-intake-builderSkill
- client-workspace-scaffolderSkill
- code-deduplicatorSkill
- cold-outreach-writerSkill
- demo-animationSkill
- demo-animation-v2Skill
- discovery-briefSkill
- em-dash-removerSkill
- etsy-listing-optimizerSkill
- follow-up-radarSkill
- github-profile-proof-updaterSkill
- github-safe-publisherSkill
- idea-inboxSkill
- implement-client-feedbackSkill
- lead-scorerSkill
- local-app-public-demoSkill
- media-spec-checkerSkill
- obsidian-daily-noteSkill
- ops-status-reportSkill
- outreach-pipeline-designerSkill
- php-form-mailerSkill
- proposal-writerSkill
- release-announcement-writerSkill
- remove-ai-image-metadataSkill
- research-brief-curatorSkill
- research-to-articleSkill
- screen-recorderSkill
- session-handoffSkill
- skill-builderSkill
- skill-reviewerSkill
- social-campaign-proof-trackerSkill
- skill-nameSkill
- thumbnail-legibility-checkSkill
- thumbnail-set-designerSkill
- vault-janitorSkill
- verify-public-claimsSkill
- web-scraperSkill
- website-change-verifierSkill
- windows-reinstall-plannerSkill
- write-evidence-backed-profileSkill
The file
We split this skill into the rules it gives an agent
10 rules. 0 are common, 6 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
Local project: if the current folder is the website's source, list the page files (HTML/PHP/markdown), read each page's <title> and meta …
- Very specific: line 17
Always exclude: form handlers and endpoints (send.php and similar), admin pages, login pages, thank-you/confirmation pages, paginated …
- Distinctive: line 21
Follow the llms.txt format:
- Distinctive: line 56
Every URL in the file must resolve (check the live ones if the user agrees, otherwise check against the sitemap).
- Very specific: line 57
Place the file at the site root as llms.txt (same level as robots.txt).
- Very specific: line 58
If the site is large, offer a second llms-full.txt with expanded per-page content.
- Distinctive: line 63
Never include pages blocked by robots.txt, behind logins, or that handle form data.
- Distinctive: line 64
Never inflate: no keyword stuffing, no claims that are not on the site.
- Distinctive: line 65
Keep the whole file under roughly 100 lines for a typical site.
- Distinctive: line 66
If the site already has an llms.txt, diff against it and report what changed rather than silently replacing.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
llms.txt Generator
You generate the llms.txt file: a markdown map of a website placed at the site root so LLMs and AI search engines can find the right pages and describe the site accurately. Think of it as a sitemap written for a reader with no patience: every page gets one honest line about what is on it.
Step 1: Inventory the site
Work from the best available source, in this order:
- Local project: if the current folder is the website's source, list the page files (HTML/PHP/markdown), read each page's
<title>and meta description, and note its purpose. - Live site: otherwise fetch the site's
sitemap.xmland read the listed pages. Respectrobots.txt: anything disallowed there stays out of llms.txt too.
Always exclude: form handlers and endpoints (send.php and similar), admin pages, login pages, thank-you/confirmation pages, paginated duplicates, and anything the owner would not want quoted.
Step 2: Write the file
Follow the llms.txt format:
# Site Name
> One or two sentences saying what the site is and who it serves. This is
> the line an AI will repeat when describing the site, so make it accurate.
Optional short paragraph with context that does not fit the summary:
what the business does, where it operates, what makes it specific.
## Main sections
- [Page Title](https://example.com/page): one line on what this page contains
- [Another Page](https://example.com/other): same again
## Resources
- [Blog](https://example.com/blog): what the blog covers
- [Downloads](https://example.com/tools): what can be downloaded
## Optional
- [Secondary pages](https://example.com/archive): things an AI can skip unless askedWriting rules for descriptions:
- Describe what is ON the page, not what the page hopes to achieve. "Pricing for the three service tiers" beats "Discover our flexible options".
- Absolute URLs throughout.
- The
## Optionalsection is for pages an AI can skip when context is tight; use it for archives and secondary material. - The summary blockquote is the highest-value line in the file. Draft it, then check: would the owner be happy if an AI quoted this verbatim as the description of their business?
Step 3: Validate and place
- Every URL in the file must resolve (check the live ones if the user agrees, otherwise check against the sitemap).
- Place the file at the site root as
llms.txt(same level asrobots.txt). - If the site is large, offer a second
llms-full.txtwith expanded per-page content; for most small business sites one file is right. - Remind the user to regenerate when pages are added or removed; a stale llms.txt misdirects the AIs it was meant to help.
Rules
- Never include pages blocked by robots.txt, behind logins, or that handle form data.
- Never inflate: no keyword stuffing, no claims that are not on the site. The file's only job is accuracy.
- Keep the whole file under roughly 100 lines for a typical site; it is a map, not a mirror.
- If the site already has an llms.txt, diff against it and report what changed rather than silently replacing.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

