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.
vitaecontextllms.txtvitaecontext/vitaecontext/llms-full.txt
install notes, manifests, command wrappers, and extension metadata. - `src/` and `bin/`: CLI and stateless MCP server engine. - `mcp/`: Model Context Protocol documentation, configs, and schema contracts. ## 5. Skill behavior `SKILL.md
vitaecontextllms.txtvitaecontext/vitaecontext/llms.txt
skill maintains an optional deeper local career knowledge graph. A stateless Model Context Protocol (MCP) server provides seamless cross-repository context access. Evidence labels and provider-aware install layouts support
adelinamartllms.txtadelinamart/robrain/llms.txt
robrain install --self-hosted` — starts Postgres + Perception in Docker and wires the MCP server into detected editors. - Register a project: `npx robrain init-project` inside the repo. Capture is automatic
akutishevskyllms.txtakutishevsky/nutrition-mcp/public/llms.txt
directory at https://claude.ai/directory/nutrition-mcp. Nutrition MCP is built as an MCP server from the ground up, so tracking lives inside the AI assistant you already use. You describe
FranciscoThiesenllms.txtFranciscoThiesen/mirror_bridge/llms-full.txt
doctor` emit exactly one JSON object on stdout, with actionable suggestions on failure - **MCP server**: gives any MCP-capable agent `generate_bindings`, `doctor`, and `check_binding_drift` as tool calls
Vortx-AIllms.txtVortx-AI/emem/web/llms.txt
with the method next to each number. - [MCP descriptor](https://emem.dev/.well-known/mcp.json): well-known MCP server descriptor for auto-discovery. Its `a2a` block is the agent-to-agent front door
vitalopsllms.txtvitalops/opendesk/llms.txt
OpenDesk > OpenDesk by VitalOps — open-source computer-use MCP server that lets AI agents see and control macOS, Windows and Linux desktops, locally or across paired machines. Python package `opendesk
jangtrinhllms.txtjangtrinh/design-os-studio/docs/llms.txt
Claude Code, Codex, Antigravity and Grok, Live Variables Mode Switching & Multi-Edit, and an MCP server. Native GPU-accelerated interface design studio (`wgpu` with a `tiny-skia` CPU renderer; Metal
jangtrinhllms.txtjangtrinh/design-os-studio/llms.txt
github.com/jangtrinh/design-os-studio/blob/main/docs/install.md) - [Prompt & CLI Cookbook](https://github.com/jangtrinh/design-os-studio/blob/main/docs/cookbook.md) - [Release & Packaging Guide](https://github.com/jangtrinh/design-os-studio/blob/main/docs/releasing.md) - [MCP Server (22 Tools)](https://github.com/jangtrinh/design-os-studio/blob/main/crates/designos-mcp/README.md)
ThreeMoonsLabllms.txtThreeMoonsLab/agents-shipgate/llms.txt
invoke models. - Does not run agents. - Does not call tools. - Does not connect to MCP servers. - Does not make scanner network calls by default. - Does not collect scanner telemetry
trustablllms.txttrustabl/agent-reliability-analyzer/llms.txt
scan, one contract, many artifacts.** A scan inventories every tool, agent, subagent, skill, and MCP server, checks them against the rule packs, and distills the result into one canonical **contract
AuthPlanellms.txtAuthPlane/authserver/llms.txt
single Go binary. Use it to add MCP-spec-compliant authorization to an MCP server, federate to an existing IdP, broker upstream provider OAuth (GitHub, Slack, Google, Linear, ...), and record
nirholasllms.txtnirholas/agenti/llms.txt
reference - [Vision](https://github.com/nirholas/agenti/blob/HEAD/README.md#vision) - [Packages](https://github.com/nirholas/agenti/blob/HEAD/README.md#packages) - [Quick Start](https://github.com/nirholas/agenti/blob/HEAD/README.md#quick-start) - [MCP Server: Plug Into Any LLM](https://github.com/nirholas/agenti/blob/HEAD/README.md#mcp-server-plug-into-any-llm) - [Framework Integrations](https://github.com/nirholas/agenti/blob/HEAD/README.md#framework-integrations) - [How Payments
HorizunGroupllms.txtHorizunGroup/horizun-revit-mcp/llms.txt
bridge. Horizun Revit MCP has a public, downloadable Windows installer. The installer includes the MCP server runtime and Revit add-ins. End users do not need Git, Visual Studio
davidmosiahllms.txtdavidmosiah/google-health-mcp/docs/llms.txt
Google Health MCP Unofficial Unofficial local-first MCP server for Google Health API v4. Important tools: google_health_connection_status, google_health_data_inventory, google_health_data_type_coverage, google
davidmosiahllms.txtdavidmosiah/google-health-mcp/llms.txt
Google Health MCP Unofficial local-first Google Health API v4 MCP server for AI health, sleep and activity agents. Repository: https://github.com/davidmosiah/google-health-mcp Homepage: https://wellness.delx.ai/connectors/google-health NPM: https://www.npmjs.com
AgentAnycastllms.txtAgentAnycast/agentanycast/llms-full.txt
traversal, peer connections, task state machine, offline message queue, BoltDB persistence, HTTP bridge, and MCP server. ## Repository Structure | Repo | Language | Purpose | |------|----------|---------| | agentanycast | Docs | Main entry, docs, issue tracker | | agentanycast-python
AgentAnycastllms.txtAgentAnycast/agentanycast/llms.txt
mcp_tool(skill) # Skill -> MCPTool card = mcp_tools_to_agent_card("WeatherServer", [tool]) # Wrap MCP server as A2A agent ``` ```typescript import { mcpToolToSkill, skillToMcpTool, mcpToolsToAgentCard } from "agentanycast"; const skill = mcpToolToSkill({ name
maximem-aillms.txtmaximem-ai/maximem_synap_sdk/llms.txt
production AI agents. Open-source Python and JS SDKs, 24 framework integrations and an MCP server; the memory engine runs as a managed service. ## Set up Synap for your user
htlin222llms.txthtlin222/openevidence-mcp/docs/llms.txt
OpenEvidence MCP > Unofficial open-source MCP server for OpenEvidence. ## What It Is - OpenEvidence is used by doctors and clinical teams for evidence-backed medical Q&A. - OpenEvidence MCP connects that
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.
