How real projects summarise themselves for language models.
NirDiamantllms.txtNirDiamant/Agentic_Engineering/llms.txt
Keeping the docs layer from drifting away from the code. - [tester](https://github.com/NirDiamant/Agentic_Engineering/tree/main/skills/tester): Test authoring support. - [testify](https://github.com/NirDiamant/Agentic_Engineering/tree/main/skills/testify): Test generation and verification. ## Evaluation - [eval](https://github.com/NirDiamant/Agentic_Engineering/tree/main/eval):
AnastasiyaWllms.txtAnastasiyaW/knowledge-space/docs/llms.txt
Complete documentation covers architecture, data models, business processes, authorization, and component relationship diagrams. - [API Testing Tools](https://happyin.space/architecture/api-testing-tools/): Practical tools for testing, debugging, and validating APIs during development
infiniVllms.txtinfiniV/ultra-instinct-claude-code/llms.txt
control what survives compaction: ## Compact Instructions When summarizing, always preserve: file paths, errors, test results, decisions, TODOs. tags for critical rules: NEVER run DROP without WHERE clause. Update CLAUDE.md after
junnam586llms.txtjunnam586/goguma/site/llms-full.txt
thing that set it disappears. ## The routes that don't work, tested It is worth recording the dead ends, because they look plausible and cost a day each. **Private
Microsoftllms.txtmicrosoft/eval-recipes/llms.txt
creating CLI applications to automations. Agents are automatically scored based on deterministic and semantic tests using a specialized auditing agent. Additional features include agent continuation (automatically providing follow-up prompts
nicholasglazerllms.txtnicholasglazer/gnosis-mcp/llms-full.txt
bash gnosis-mcp ingest ./docs/ # Auto-creates DB + loads markdown gnosis-mcp search "query" # Test it works gnosis-mcp serve # Start MCP server ``` ## Quick Setup (SQLite + Semantic Search) ```bash
nicholasglazerllms.txtnicholasglazer/gnosis-mcp/llms.txt
stdio protocol: 8.7 ms mean, 13.0 ms p95 (v0.13.3 SDK 1.27 transport). 862 tests, 10 RAG eval cases (Hit@5 = 1.00, MRR = 1.00, Precision@5 = 0.75). Install size: ~23MB with
redis-developerllms.txtredis-developer/adk-redis/docs/llms.txt
only/REPOSITORY_MAP.md): source tree and where each feature lives. - [Build and test](for-ais-only/BUILD_AND_TEST.md): make targets, unit and integration loops, mkdocs build. - [Failure modes](for-ais-only/FAILURE_MODES.md
shaqmughalllms.txtshaqmughal/seekstone/llms.txt
pinned). - **Write safety:** Seekstone ships a named Write-Safety Contract (docs/WRITE-SAFETY.md) of ten tested guarantees: zero network/telemetry, vault sandbox, byte-identical frontmatter on body edits, atomic writes (no torn files
Bordallms.txtBorda/AI-Rig/docs/llms.txt
Codemap-py](https://borda.github.io/AI-Rig/codemap-py/): 6 dual-runtime skills for static Python indexing, queries, test impact, renames, integration, and telemetry debriefs - [Codex Rig](https://borda.github.io/AI-Rig/codex-rig/): 13 evidence-first workflows
jedardenllms.txtjedarden/NEEDLE/llms.txt
bead init --prefix ## Create first bead bead create \ --title 'Add a CONTRIBUTING.md so that test -s CONTRIBUTING.md passes. Work on this issue directly.' \ --description 'Create CONTRIBUTING.md with contribution guidelines. Acceptance
michielhdotethllms.txtmichielhdoteth/awesome-ai-agent-tools/llms.txt
categories (Development 35, Productivity 18, Content 13, Design 11, DevOps 8, Marketing 6, Testing 4, Data 4, Security 2) - **140 MCP Servers**: Model Context Protocol servers with install commands across
zinimanllms.txtziniman/ai-instruct/llms.txt
/ziniman/ai-instruct/main/web-accessibility-guide.md): WCAG 2.2 AA — semantic HTML, keyboard navigation, ARIA, forms, color contrast, focus management, testing. - [Web Performance](https://raw.githubusercontent.com/ziniman/ai-instruct/main/web-performance-guide.md): Core Web Vitals (LCP, CLS, INP), image and font optimization
Foxfire1stllms.txtFoxfire1st/agents-remember/docs/llms.txt
registry of where the project's documentation lives, the exact lint / type-check / test / build commands for the repo, its coding guidelines, its branch-and-landing (`git-workflow`) flow
IgorGanapolskyllms.txtIgorGanapolsky/ThumbGate/.well-known/llms.txt
covering one supported workflow — a 60-minute review, one configured local gate, its regression test, and written rollout/rollback proof within two business days. Current terms: https://thumbgate.ai/pricing - Enterprise: custom
tcsenpaillms.txttcsenpai/specification-website-skill/references/llms-full.txt
Check the `Content-Type` response header in the Network tab. - Add a non-ASCII test string (`café — 日本語 — 🌍`) to a page and confirm it renders correctly. ### Sources - HTML Living Standard
wazionappsllms.txtwazionapps/nexo/.well-known/llms.txt
connection closes, SQLite close is serialized under the write lock, and the full Brain test workflow has enough time to finish instead of cancelling slow-but-valid runs. Builds
wundercorpllms.txtwundercorp/loki/llms-full.txt
need — it won't touch or interfere with any system Git. > **Android / Termux:** The tested manual path is documented in the [Termux guide](https://loki.computer/docs/getting-started/termux). On Termux, Loki installs
zircotellms.txtzircote/.claude/llms.txt
deployment-engineer, network-engineer - security-engineer, incident-responder, devops-incident-responder - `04-quality-security/` - Testing, security, review (11 agents) - qa-expert, test-automator, code-reviewer - security-auditor, penetration-tester, compliance
myclaude-shllms.txtmyclaude-sh/myclaude-creator-engine/llms.txt
Profile:** `/onboard` — engine adapts to their expertise and technical level 4. **Research:** `/scout ` — tests Claude's baseline knowledge, finds gaps, scans marketplace 5. **Create:** `/create ` — generates structured product with quality
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.
