How real projects summarise themselves for language models.
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.
Kanevryllms.txtKanevry/session-orchestrator/site/llms.txt
orchestrator@kanevry` (afterwards run `npm install` once inside the plugin directory and restart Claude Code) - Upgrade: `/plugin update session-orchestrator@kanevry`, then restart. Session-start warns when the RUNNING copy
Rtur2003llms.txtRtur2003/Claude-Code-Promts-Skills/llms.txt
/prompts/english/project-types/blockchain-web3-prompt.md - Language-agnostic: ./prompts/english/project-types/general-software-development-prompt.md - Foundation (universal best practices): ./prompts/english/base/claude-foundation-prompt.md ## Improving existing code (specialist prompts) - Code review: ./prompts/english/agents/code-review-prompt.md - Security audit: ./prompts/english/agents/security-audit-prompt.md - Debugging / incident: ./prompts/english/agents/debugging-troubleshooting-prompt.md - Refactoring: ./prompts/english/agents/refactoring-prompt.md - Testing strategy: ./prompts/english/agents/testing-strategies-prompt.md - Performance: ./prompts/english/agents/performance-optimization-prompt.md
antongulinPipes a download into a shellllms.txtantongulin/robin/llms.txt
Robin > Free AI code reviews on every pull request. Robin runs as a GitHub Action in your own > repository, calls an OpenAI-compatible LLM endpoint you configure (OpenRouter's free
danicatllms.txtdanicat/speedgrapher/llms.txt
Models (LLMs), AI search engines, and coding agents. ## Core Documentation - [README.md](README.md): Architecture overview, installation guide, four-mode sidecar panel walkthrough (`Reader`, `Editor`, `Reviewer`, and `Raw`), MCP server tools
xiaobei930llms.txtxiaobei930/cc-best/docs/llms.txt
skills (backend, frontend, testing, security, architecture, devops, git, learning) - 8 agents (architect, planner, code-reviewer, code-simplifier, security-reviewer, tdd-guide, build-error-resolver, requirement-validator) - 43 rules across
almogdepazPipes a download into a shellllms.txtalmogdepaz/wolfpack/site/llms-full.txt
serve` are verified does setup also print a private Tailnet HTTPS URL and QR code. If you decline the login service, run `wolfpack` before opening the printed URL. Use `wolfpack
Naridon-IncPipes a download into a shellllms.txtNaridon-Inc/aura/llms.txt
intent`, `aura_snapshot`, `aura_rewind`, `aura_pr_review`, `aura_prove` and `aura_usage`. - Sessions are detected and tracked from Claude Code, Gemini CLI, Codex CLI, Kimi, Cursor and OpenCode
pedrobalsallms.txtpedrobalsa/balsa-ui/public/llms-full.txt
palette trigger with a custom saturation, hue, and editable HEX, RGB, or HSL color-code popover. - [Modal](https://balsa-ui.com/docs/components/modal.md): Typed semantic-color surface, solid, outline, soft, or glass focused
wandercomllms.txtwandercom/kindex/docs/llms-full.txt
candidate accept --review-token \ --by reviewer --method source-check kin candidate reject --by reviewer --code not_relevant kin candidate prune kin candidate erase ``` Verification is typed provenance, not a title
wandercomllms.txtwandercom/kindex/docs/llms.txt
Kindex Kindex is a local-first persistent knowledge graph and MCP server that Claude Code, Codex, Gemini CLI, Google Antigravity, OpenCode, Cursor, and other MCP-capable agents all read
SannePipes a download into a shellllms.txtSanne/incus-spawn/site/llms.txt
remote isx manages, so you review an immutable commit, never a directory the agent can still change. isx works with Claude Code, Codex CLI, Pi, GitHub Copilot CLI and other
video-dbPipes a download into a shellllms.txtvideo-db/agent-toolkit/context/llms-full.txt
# Task --- You are an advanced assistant specialized in generating VideoDB Python code. You have
zuke-buildReads credentialsllms.txtzuke-build/zuke/llms-full.txt
# Zuke — full API reference The complete typed public surface of every package. Each section
zuke-buildllms.txtzuke-build/zuke/llms.txt
task wrappers for free, open-source security scanners - [@zuke/ai](https://jsr.io/@zuke/ai) — AI-powered code review for Zuke builds - [@zuke/otel](https://jsr.io/@zuke/otel) — an OpenTelemetry (OTLP/HTTP JSON) export plugin
codesodaPipes a download into a shellllms.txtcodesoda/discuss-cli/llms.txt
preserve edits; success emits `session.done`. Standalone PR comments are out of scope. - When the review finishes, discuss emits `session.done` with the final transcript payload. With `--verdict-options`, finishing requires
memi-designllms.txtmemi-design/memi/llms.txt
mappings are a fallback when native Code Connect is unavailable. Validate mappings against code; reuse the project's existing components and styling conventions. Do not force a shadcn migration
Agnuxo1llms.txtAgnuxo1/OpenCLAW-P2P/llms.txt
agent entering the P2PCLAW network — a decentralized platform where AI agents publish and peer-review scientific papers. Your task: register, pass the Tribunal exam, write a research paper, and publish
jmrplensllms.txtjmrplens/gitlab-mcp-server/llms-full.txt
# gitlab-mcp-server. Full Reference > Version 3.4.0 | up to 1099 tools | 34 base meta
jmrplensReads credentialsllms.txtjmrplens/gitlab-mcp-server/llms.txt
resources/subscribe on stateful sessions; change detection honored by polling): - gitlab://groups: groups - gitlab://guides/code-review: code_review - gitlab://guides/conventional-commits: conventional_commits - gitlab://guides/git-workflow: git_workflow - gitlab://guides/merge-request-hygiene: merge_request_hygiene
mshadmanrahmanReads credentialsllms.txtmshadmanrahman/claudecode-guide/public/llms-full.txt
Claude Code instance that handles a focused task. Your main conversation delegates work (research, code review, testing) to agents that work independently and report back. Keeps your main session clean
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.
