How real projects summarise themselves for language models.
GuyMannDudellms.txtGuyMannDude/mnemo-cortex/llms.txt
github.com/GuyMannDude/mnemo-cortex/tree/master/integrations/ollama-desktop): Terminal launcher with OpenClaw as MCP host. ## Agent Manifests - [INSTALL.md](https://github.com/GuyMannDude/mnemo-cortex/blob/master/INSTALL.md): The install guide, in plain English — server setup plus per-host wiring for Claude Desktop
pc-stylellms.txtpc-style/x-md/public/llms.txt
contract: `Accept: application/json` on any read route, described by https://x.pcstyle.dev/openapi.json, or the MCP server at https://x.pcstyle.dev/mcp. How to call it: one plain
ayush-thatllms.txtayush-that/sub-agents.directory/public/llms.txt
Agents Directory > A community-driven directory of Claude Code sub-agents, rules, prompts, and MCP servers. Sub-Agents Directory is the go-to resource for developers using Claude Code
Cassette-Editorllms.txtCassette-Editor/oh-my-cassette/llms.txt
Cassette > Oh My Cassette is an open-source AI video editing plugin and local MCP server for Claude Code, Codex, Hermes Agent, and OpenCode. It turns natural-language chat into
KryptosAIllms.txtKryptosAI/mcp-observatory/llms.txt
Observatory is a free, local-first CI and security gate for Model Context Protocol (MCP) servers before AI agents depend on them. ## Use This When - You maintain an AI agent
JasonColapietrollms.txtJasonColapietro/suede-creator-skills/docs/llms.txt
grade, strip the slop out of the copy, then ship. Also covers the MCP server, install paths, and the FAQ; the complete list is on the skill index above. - [Skill
hidai25llms.txthidai25/eval-view/llms-full.txt
Responses API) | Yes | Yes | | Anthropic Claude | Yes | Yes | | HuggingFace | Yes | Yes | | Ollama | Yes | Yes | | MCP servers | Yes | Yes | | Any HTTP API | Yes | Yes | | Claude Code (skills) | Yes | Yes | | OpenAI Codex
hidai25llms.txthidai25/eval-view/llms.txt
affiliated with them. ## Supported Frameworks LangGraph, CrewAI, OpenAI (Responses API), Anthropic Claude, HuggingFace, Ollama, MCP servers, and any HTTP API. ## Documentation - Getting Started: https://github.com/hidai25/eval-view/blob/main/docs/GETTING_STARTED.md - CLI Reference: https://github.com
deepnoodle-aillms.txtdeepnoodle-ai/dive/llms.txt
dive.ToolAdapter(&WeatherTool{DB: db}) ``` ## Dynamic tools (Toolset) Use `Toolset` for tools resolved at runtime (MCP servers, permission-filtered, etc.): ```go agent, _ := dive.NewAgent(dive.AgentOptions{ Model: anthropic.New(), Toolsets: []dive.Toolset{ &dive.ToolsetFunc{ ToolsetName: "mcp
benseverndev-ossllms.txtbenseverndev-oss/goldenmatch/llms.txt
stable entity_ids that survive across runs) ## AI-native surface - Every package ships an MCP server (139 tools across the suite), a REST API, and most an A2A agent surface
ChanMeng666llms.txtChanMeng666/server-google-news/llms.txt
Google News MCP Server - AI Usage Guide ## Service Overview This is a Model Context Protocol (MCP) server specifically designed to provide Google News search capabilities for AI agents. ## Core Features
Bande-a-Bonnotllms.txtBande-a-Bonnot/Boucle-framework/docs/llms.txt
verified root after install, upgrade, restore, or repair. - After Claude Code, hooks, settings, MCP servers, connectors, or plugins change, capture `--verify --summary-only` from the same project root first, then
chattermatellms.txtchattermate/chattermate.chat/llms.txt
reference](https://docs.chattermate.chat/features/cli): the account CLI — signup, login, tokens, agents, workflows, knowledge. - [MCP Server](https://docs.chattermate.chat/features/mcp-server): let an AI agent configure ChatterMate over the Model Context Protocol (`chattermate
sandbaseaillms.txtsandbaseai/deepseek-harness-handbook/site/llms.txt
DNS/TCP transport, and provider HTTP responses without disabling TLS. - [DeepSeek Harness MCP guide](https://sandbaseai.github.io/deepseek-harness-handbook/deepseek-harness-mcp.html): Add one server to a user Agent preset, start a fresh Session generation
X-isdoingreatllms.txtX-isdoingreat/canvas-pilot/site/llms.txt
students whose schools use Canvas LMS. - A workflow layer above Canvas access and Canvas MCP servers. - A way to reuse course-specific recurring assignment patterns week after week. - A read
OriNachumllms.txtOriNachum/claude-code-guide/llms.txt
Configuration & Extensions]({{ site.url }}/configuration.md): Memory, settings, plugins, the marketplace, and MCP. - [Integrations]({{ site.url }}/integrations.md): MCP servers and connecting Claude Code to other tools. - [User Stories]({{ site.url }}/user-stories.md): Narrative walkthroughs
sinewaveaillms.txtsinewaveai/agent-security-scanner-mcp/llms.txt
agent-security-scanner-mcp > MCP server providing real-time security scanning for AI coding agents. Detects vulnerabilities, hallucinated packages, and prompt injection across 12 languages. ## Tools - scan_security: Scan
thebanrillms.txtthebanri/limoni/llms.txt
style tests against it by role and label, and `limoni-mcp` serves it to MCP clients such as Claude Code and Cursor, so an agent can operate a running
agents-uillms.txtagents-ui/agents-kit/llms-full.txt
Generative playground](https://agents-ui.github.io/agents-kit/generative) - [MCP setup](https://agents-ui.github.io/agents-kit/docs/mcp): use the standard shadcn MCP server with the @agents-kit registry namespace. - [Registry index](https://agents-ui.github.io/agents-kit/c/registry.json) - [Full API reference
agents-uillms.txtagents-ui/agents-kit/llms.txt
Generative playground](https://agents-ui.github.io/agents-kit/generative) - [MCP setup](https://agents-ui.github.io/agents-kit/docs/mcp): use the standard shadcn MCP server with the @agents-kit registry namespace. - [Registry index](https://agents-ui.github.io/agents-kit/c/registry.json) - [Full API reference
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.
