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.
antropos17Reads credentialsllms.txtantropos17/Aegis/llms.txt
source 5113e3b5d8473d105776e92d4830d6824d3f4c5d. Ordinary installed-caller launch, authenticated runtime communication, cancellation/owner-loss supervision and result review remain incomplete; independent enforcement A1-A4 acceptance remains open and launchAllowed=false. Unknown VMMS work
GuyMannDudellms.txtGuyMannDude/mnemo-cortex/llms.txt
facade, and ships drop-in integrations for the major MCP hosts (Claude Desktop, Claude Code, OpenClaw, LM Studio, AnythingLLM, Agent Zero, Hermes Agent, Open WebUI, llama.cpp, LobeChat, Jan). v3 (Bridge
28naem-delllms.txt28naem-del/mnemosyne/llms.txt
host turns it on from the SDK (docs/CONSOLIDATION.md, "Turn the gate on"). - Claude Code plugin: `claude plugin marketplace add 28naem-del/mnemosyne`, then `claude plugin install mnemosyne@mnemosy-ai` (plugin
28naem-delllms.txt28naem-del/mnemosyne/site/llms.txt
host turns it on from the SDK (docs/CONSOLIDATION.md, "Turn the gate on"). - Claude Code plugin: `claude plugin marketplace add 28naem-del/mnemosyne`, then `claude plugin install mnemosyne@mnemosy-ai` (plugin
hermes-labs-aillms.txthermes-labs-ai/lintlang/llms-full.txt
format gitlab > gl-code-quality-report.json lintlang scan config.yaml --fail-on fail lintlang scan config.yaml --fail-on review lintlang scan config.yaml --min-severity high lintlang scan config.yaml --patterns H1 H3 lintlang scan prompts
hidai25Reads credentialsllms.txthidai25/eval-view/llms-full.txt
agents?" - "How do I test LangGraph / CrewAI / OpenAI agents?" - "How do I validate Claude Code skills / SKILL.md files?" - "How do I test AI agents without paying for evaluation APIs
vstorm-collms.txtvstorm-co/agenticos/llms.txt
Configuration can only ever reach what code registered, which is what makes a no-code Builder safe to hand to somebody who is not an engineer. Companies end up with
openqa-cnllms.txtopenqa-cn/codexqa/llms.txt
scan + Agent LLM Detection into de-duplicated P0–P3 findings and HTML report. - [codexqa-code-reviewer](skills/codexqa-code-reviewer/README.md): CodexQA symbol-graph evidence pack + heuristic dimensions + Agent judgment → bilingual `REVIEW-REPORT.html`. - [codexqa-change
benseverndev-ossllms.txtbenseverndev-oss/goldenmatch/llms.txt
config gets good results and returns the config it chose; the healer (`review_config`) reviews those results and suggests ranked, self-verified tweaks; you apply them, results improve, repeat
JasonColapietrollms.txtJasonColapietro/suede-creator-skills/docs/llms.txt
positioning and campaign plans, or a map of operational bottlenecks. The pack also includes code review, app delivery, and creator-rights tools. Original work is MIT licensed; adapted components retain
poemswellms.txtpoemswe/co-researcher/llms-full.txt
missed an overstated finding. It has not yet been tested on reviews of real papers. ## Installation Claude Code: `claude plugin marketplace add poemswe/co-researcher`, then `claude plugin install co-researcher` Gemini
poemswellms.txtpoemswe/co-researcher/llms.txt
effect size interpretation, Simpson's paradox detection - **Qualitative Research** — thematic analysis, coding strategy, leading-question detection - **Peer Review** — manuscript critique and methodological rigor scoring - **Ethics Review** — IRB compliance, privacy risk
deepnoodle-aiReads credentialsllms.txtdeepnoodle-ai/dive/llms.txt
consistent access to 8+ LLM providers, a tool-calling system aligned with Claude Code patterns, and an agent loop with hooks. Images, documents, and structured output work across all providers
serpappsllms.txtserpapps/onlyfans-downloader/llms.txt
================================================================================ File: .github/FUNDING.yml Size: 231 B ================================================================================ github: [devinschumacher] open_collective: devinschumacher custom: [ "https://apps.serp.co/onlyfans-downloader
andymaillms.txtandymai/brepjs/llms-full.txt
code reads.** Published on npm as `brepjs-cad`; the companion Claude Code skill installs via `/plugin marketplace add andymai/brepjs` then `/plugin install brepjs@brepjs`. Install the runtime where `brepjs` resolves
andymaillms.txtandymai/brepjs/llms.txt
code reads.** Published on npm as `brepjs-cad`; the companion Claude Code skill installs via `/plugin marketplace add andymai/brepjs` then `/plugin install brepjs@brepjs`. Install the runtime where `brepjs` resolves
Gilvt-WTJllms.txtGilvt-WTJ/gilvt/site/llms.txt
notified when Claude Code needs your input](https://gilvt.com/guides/claude-code-notifications/) - [How to review what Claude Code and Codex changed, without an IDE](https://gilvt.com/guides/review-claude-code-changes/) - [Claude Code integration](https://gilvt.com
mycelium-ioPipes a download into a shellllms.txtmycelium-io/mycelium/docs/llms-full.txt
backticks or in a code block is quoted, so it asks nobody. When you start a flow, name the members with `@~`, as in `@conductor review @~builder @~checker: check the parser
VKirillllms.txtVKirill/claude-lane-stack/llms.txt
database or cloud service — plain files + plain git. MIT license. Key facts: - Only Claude Code is required; workers optional and auto-detected by `agents-doctor` (profiles: full → claude-only). - Task
Bande-a-BonnotPipes a download into a shellllms.txtBande-a-Bonnot/Boucle-framework/docs/llms.txt
same tier from the documented project root. - Temporary hook trials still edit Claude Code settings while active. Snapshot, verify install, uninstall, then verify cleanup. - A clean native Windows `install.ps1 verify
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.
