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.
melgarafaelllms.txtmelgarafael/DeskcommCRM/public/llms.txt
budget; self-improvement loop (resolved conversations feed the knowledge base) - MCP: internal MCP server today; public MCP contract on the roadmap — any AI agent can operate the CRM - WhatsApp
Steady5 mentions in 2 weeks
No other project uses it yet
DavidHDevllms.txtDavidHDev/canvas-ui/public/llms.txt
WebGL and WebGPU (vgpu) builds, browser support, and how the WebGPU build works - [MCP Server](https://canvasui.dev/docs/mcp): Use the registry with the shadcn MCP server so AI assistants
No other project uses it yet
eugeniughelburReads credentialsllms.txteugeniughelbur/obsidian-second-brain/llms.txt
recall, every command real-tested; ships the Hermes native-skills build and the MCP vault server. https://github.com/eugeniughelbur/obsidian-second-brain/releases/tag/v0.11.0 - v0.10.0 (2026-05-31) "The Architect": `/obsidian-architect` scans a codebase into
mock-serverllms.txtmock-server/mockserver/jekyll-www.mock-server.com/llms-full.txt
MockServer > MockServer is an open-source HTTP(S) mock server, recording proxy, and **MCP (Model Context Protocol) server for AI coding assistants**. It enables easy mocking of any system
No other project uses it yet
mock-serverllms.txtmock-server/mockserver/jekyll-www.mock-server.com/llms.txt
streams — no code changes - [AI Protocol Mocking (MCP & A2A)](https://www.mock-server.com/mock_server/ai_protocol_mocking.html): Mock MCP servers and A2A agent-to-agent endpoints (over SSE, WebSocket, and JSON-RPC) to test
No other project uses it yet
langwatchReads credentialsllms.txtlangwatch/langwatch/docs/llms-full.txt
project id in the environment. The SDKs and the MCP server read both. ```bash .env LANGWATCH_API_KEY="sk-lw-..." LANGWATCH_PROJECT_ID="your-project-id" ``` The project
No other project uses it yet
langwatchllms.txtlangwatch/langwatch/docs/llms.txt
assistant. Instrument code, version prompts, run scenarios, inspect traces, query analytics, and more. - [LangWatch MCP Server](https://langwatch.ai/docs/integration/mcp.md): Use the LangWatch MCP Server to extend your coding assistant with
No other project uses it yet
NewFuturellms.txtNewFuture/DDNS/docs/llms.txt
ddns.newfuture.cc/release.md)) ### Configuration Guides - CLI Arguments: https://ddns.newfuture.cc/config/cli ([.md](https://ddns.newfuture.cc/config/cli.md)) - MCP Server: https://ddns.newfuture.cc/config/mcp ([.md](https://ddns.newfuture.cc/config/mcp.md)) - Environment Variables: https://ddns.newfuture.cc/config/env ([.md](https://ddns.newfuture.cc
No other project uses it yet
dagucloudReads credentialsllms.txtdagucloud/dagu/llms.txt
server (list, create, run, stop, schedule), follow `references/server-operations.md`: detect whether a server exists without asking, then prefer MCP, then REST, then a CLI remote context, and use the local
dagu-orgReads credentialsllms.txtdagu-org/dagu/llms.txt
Python or PowerShell script. Inspect a workbook first with `dagu xlsx inspect` or the MCP `workbook` read target. - Prefer `git.worktree.add` and `git.worktree.remove` when steps need isolated branches inside an existing
1 project uses a version of this
openagents-orgReads credentialsllms.txtopenagents-org/openagents/llms.txt
MyNetwork mode: centralized transports: - type: http config: port: 8700 serve_mcp: true # Expose as MCP server - type: grpc config: port: 8600 mods: - name: openagents.mods.workspace.messaging enabled: true config: default_channels: - name
No other project uses it yet
flypythoncomllms.txtflypythoncom/python/llms.txt
stateless 2026-07-28 MCP tool server in pure Python — reuses the reviewed mcp-server example core (EN+ZH). - `courses/da-eda/`: Exploratory data analysis with pandas over a messy shop export
No other project uses it yet
txn2llms.txttxn2/kubefwd/docs/llms-full.txt
kubefwd.internal/api` ## Key Endpoints ### Health & Info - `GET /api/health` - Health check - `GET /api/info` - Server information ### Services - `GET /api/v1/services` - List forwarded services - `GET /api/v1/services/:key` - Get specific service - `POST /api/v1/services` - Add service
No other project uses it yet
steveseguinllms.txtsteveseguin/vdo.ninja/llms.txt
self-hosting project. Read its limitations; the wrapper is not itself an SFU/media distribution server. - [Broadcast-box with VDO.Ninja support](https://github.com/steveseguin/broadcast-box-with-vdon-support): Experimental fork demonstrating a self-hosted Meshcast-like
No other project uses it yet
langroidReads credentialsllms.txtlangroid/langroid/llms.txt
changes to how `FastMCPServer` forwards image context and resources, and added optional persistence for MCP server connections. The handler parameter analysis improvements were made to support more robust type checking
No other project uses it yet
datamodel-code-generatorllms.txtdatamodel-code-generator/datamodel-code-generator/docs/llms-full.txt
models from JSON Schema](https://github.com/topoteretes/cognee/blob/main/cognee/shared/graph_model_utils.py)* - [e2b-dev/E2B](https://github.com/e2b-dev/E2B) - *[Generate MCP server TypedDict models via Makefile](https://github.com/e2b-dev/E2B/blob/main/packages/python-sdk/Makefile)* - [DataDog/integrations-core](https://github.com/DataDog/integrations-core) - *[Config models
datamodel-code-generatorllms.txtdatamodel-code-generator/datamodel-code-generator/docs/llms.txt
from MCP Tool Schemas](https://datamodel-code-generator.koxudaxi.dev/mcp-tools/): Generate Python models from Model Context Protocol (MCP) tool schema profiles. - [Kubernetes Schemas](https://datamodel-code-generator.koxudaxi.dev/kubernetes/): Kubernetes describes its resources with OpenAPI schemas
koxudaxillms.txtkoxudaxi/datamodel-code-generator/docs/llms-full.txt
models from JSON Schema](https://github.com/topoteretes/cognee/blob/main/cognee/shared/graph_model_utils.py)* - [e2b-dev/E2B](https://github.com/e2b-dev/E2B) - *[Generate MCP server TypedDict models via Makefile](https://github.com/e2b-dev/E2B/blob/main/packages/python-sdk/Makefile)* - [DataDog/integrations-core](https://github.com/DataDog/integrations-core) - *[Config models
No other project uses it yet
koxudaxillms.txtkoxudaxi/datamodel-code-generator/docs/llms.txt
from MCP Tool Schemas](https://datamodel-code-generator.koxudaxi.dev/mcp-tools/): Generate Python models from Model Context Protocol (MCP) tool schema profiles. - [Generate from JSON / YAML / CSV Data](https://datamodel-code-generator.koxudaxi.dev/jsondata/): Generate Pydantic models
Steady5 mentions in 2 weeks
1 project uses a version of this
cosmicstack-labsPipes a download into a shellllms.txtcosmicstack-labs/mercury-agent/docs/llms-full.txt
Scheduled tasks and event response; persistent execution | Workspace, Microsoft 365, Slack, Salesforce, Git, Jira, MCP servers | Sub-agents and coworker agents | Identity, policy, permissions, sandboxing, network gateways | Gemini and Claude
No other project uses it yet
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.
