How real projects summarise themselves for language models.
decionisllms.txtdecionis/agent-safe-pipeline/llms.txt
authority boundary in front of an existing agent, API or service without changing its code -> `agentsafe proxy --upstream ` (Homebrew, Linux packages, Docker, Kubernetes, all the same runtime): https://github.com/decionis/agent-safe-pipeline/blob/master/docs/quickstart/README.md.
kydlikebtcllms.txtkydlikebtc/awesome-jev/llms.txt
metadata snapshots. Human review: `evidence.path` cites the file reviewed, and `evidence.read_on` records the reported reading date. Reading code is separate from running it. Summaries include source-project descriptions
ZeroPointRepollms.txtZeroPointRepo/awesome-hermes-skills/llms.txt
repository link, the author and a one line description. Install commands sit in fenced code blocks. ## Files - [README.md](https://raw.githubusercontent.com/ZeroPointRepo/awesome-hermes-skills/main/README.md): the whole list. Headings cover the skills built into
AnxForeverllms.txtAnxForever/stylekit/public/llms.txt
mode, and Tailwind UI. - [中文 UI 与前端提示词](https://www.stylekit.top/zh/ui-prompts): 面向 ChatGPT、Claude、Claude Code 与 Codex 的提示词入口。 - [Page templates](https://www.stylekit.top/en/templates): Browse complete page structures without replacing
nodetool-aillms.txtnodetool-ai/nodetool/docs/llms.txt
Building Mini Apps]({{ site.url }}/mini-apps-guide): Recipes for generators, streaming apps, galleries, live tuning, and review flows. - [Mini App Reference]({{ site.url }}/mini-apps-reference): Widgets, binding grammar, actions, conditions, and the document schema
agnt-ggllms.txtagnt-gg/agnt/docs/llms-full.txt
# AGNT API Reference — llms.txt # Auto-generated flattened API documentation for LLM consumption # Source: _API-DOCUMENTATION.md
agnt-ggllms.txtagnt-gg/agnt/docs/llms.txt
filtering, webhooks, error codes, and SDK support. - [Remote API Endpoints](https://agnt.gg/docs/api/remote-endpoints): Complete endpoint reference for the Remote API. Covers Marketplace (browse, publish, install, reviews, analytics), Users & Subscriptions (register
danielrosehillllms.txtdanielrosehill/Claude-Code-Projects-Index/docs/llms.txt
ship releases. - [Dev Tools](https://github.com/danielrosehill/Claude-Dev-Tools-Plugin): Claude Code plugin: dev-tools workflow — scaffold repos, multi-agent QA review, templatize. Session-handover commands and agent moved to claude-hopper
oxbshwllms.txtoxbshw/LLM-Agents-Ecosystem-Handbook/llms.txt
coding_agents/safe_refactoring.md), [coding_agents/review_checklist.md](coding_agents/review_checklist.md) - [coding_agents/prompts/](coding_agents/prompts/): repo audit, modernization, feature, bugfix, provider expansion, docs update, release review ## Design docs - [design_docs/README.md](design_docs/README.md) - [design_docs/agent_design_doc.md](design
Ikalus1988llms.txtIkalus1988/MisakaNet/docs/llms-full.txt
gateway-incompatible-with-anthropic-format-requires-openai-proxy](https://misakanet.org/lessons/internal-gateway-incompatible-with-anthropic-format-requires-openai-proxy/) - [Lesson: js-dead-code-chain-break](https://misakanet.org/lessons/js-dead-code-chain-break/) - [Lesson: json-解析失败handling-截断-格式error](https://misakanet.org/lessons/json-解析失败handling-截断
codexstar69llms.txtcodexstar69/bug-hunter/llms-full.txt
Hunter is a measurable, precision-first skill for AI coding agents. It combines deterministic scoping with adversarial model review, exact source integrity, bounded retrieval, optional hybrid verification, and explicit mutation
codexstar69llms.txtcodexstar69/bug-hunter/llms.txt
call out every manual-review or unreviewed item. ``` The `bug-hunter` terminal command installs and verifies the skill. Scans are requested inside the coding agent. There is no `bug-hunter
saltbollms.txtsaltbo/agent-kanban/public/llms-full.txt
assigned to a Realmroot Agent actor 3. In Progress after claim 4. In Review after submission 5. Done after a different verified actor accepts it, or In Progress after that
juyterman1000llms.txtjuyterman1000/entroly/docs/llms.txt
evident receipts, and opt-in content-addressed reuse of unchanged graph analyses. - [39-dimension code-intelligence conformance protocol](https://github.com/juyterman1000/entroly/blob/main/benchmarks/CODE_INTELLIGENCE_CONFORMANCE.md): Structural correctness, typed dispatch, repository understanding, freshness, LSP, architecture
Kevin-Liu-01llms.txtKevin-Liu-01/Claude-of-Tanks/public/llms-full.txt
driven effects. Published screenshots and videos are in-engine captures used for presentation and review; they do not replace runtime or test evidence for a technical claim. ## Source hierarchy
Kevin-Liu-01llms.txtKevin-Liu-01/Claude-of-Tanks/public/llms.txt
main-first release practice. - [Procedural model pipeline](https://cot.kevinliu.studio/docs/models): Reference study, first-party coded geometry, articulated rigs, combat anatomy, generated icons and technical cards, and visual graduation gates. - [Simulation
1mancompanyllms.txt1mancompany/OneManCompany/docs/llms.txt
cost tracking - **1-on-1 Coaching**: Shape employee behavior through direct guidance - **File Approval**: Review and approve employee file changes ## Documentation - Full docs: https://1mancompany.github.io/OneManCompany/docs/ - Getting started: https://1mancompany.github.io
Keesan12llms.txtKeesan12/martin-loop/llms-full.txt
does not claim to replace Git, GitHub, CI/CD, dedicated security scanners, observability platforms, human code review, or all testing infrastructure. It can connect work that uses those systems into
Keesan12llms.txtKeesan12/martin-loop/llms.txt
needs to work across multiple coding agents or hosts - non-engineers need a clearer path from product intent to software work that can be reviewed and trusted Technical controls such
jamditisllms.txtjamditis/claude-skills-journalism/docs/llms.txt
agents, independent tasks across 2+ agents - executing-plans, run a written implementation plan - requesting-code-review, verify completed work before merging - receiving-code-review, handle review feedback effectively - verification-before
A markdown file at a website's root that gives language models a short guide to the site and links to the pages worth reading.
The same idea with the content included, so an agent can read the documentation in one request.
Agents and tools that fetch documentation on someone's behalf. It's a proposal, not a standard, and support varies.
Start with a one-line summary, then sections of links with a sentence each. The examples here show what real projects do.
