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.
hareharePipes a download into a shellllms.txtharehare/mq/docs/llms-full.txt
Exit app | | `↑`/`k`, `↓`/`j` | Navigate results | | `d` | Toggle detail view | | `?` / `F1` | Show help screen | --- # MCP Server The mq MCP server exposes four tools for AI applications: - `html_to_markdown` - Converts
hareharellms.txtharehare/mq/docs/llms.txt
/playground): Interactive browser playground - [TUI](https://mqlang.org/book/start/tui): Interactive terminal UI for querying Markdown - [MCP Server](https://mqlang.org/book/start/mcp): Model Context Protocol integration for AI applications ## Optional - [Web Crawler
borgheiReads credentialsllms.txtborghei/Claude-Skills/site/llms.txt
when the user asks to "estimate LLM costs", "count tokens in prompts", "optimize prompt... - [Mcp Server Builder](https://borghei.github.io/Claude-Skills/skills/engineering/mcp-server-builder.html): Build MCP (Model Context Protocol) servers with tool definitions, resource
Steady6 mentions in 2 weeks
No other project uses it yet
agentic-communityPipes a download into a shellllms.txtagentic-community/mcp-gateway-registry/docs/llms.txt
Summary This project provides an enterprise-ready gateway and registry for Model Context Protocol (MCP) servers, enabling centralized management, secure access, and dynamic tool discovery for AI agents and development
sirkirbyllms.txtsirkirby/unifi-network-mcp/docs/llms.txt
UniFi MCP > UniFi MCP is an open-source collection of local-first MCP servers and an independent HTTP API for controlled access to UniFi Network, Protect, and Access infrastructure. UniFi
MartinDelophyllms.txtMartinDelophy/ai-video-editor/public/llms-full.txt
STDIO MCP operate on project files. The website does not host a remote MCP or OAuth server. - Tool results can include project metadata, caption text and explicitly requested rendered media
MartinDelophyllms.txtMartinDelophy/ai-video-editor/public/llms.txt
Tools do not provide arbitrary URL imports, media upload, remote generation or a hosted MCP server. ## Citation guidance Use the canonical pages above for current product claims. Do not describe
Vrun-designllms.txtVrun-design/openflowkit/public/llms.txt
github.com/openflowkit/openflowkit Write Mermaid or OpenFlow DSL, hand it to the `openflowkit` MCP server, and save the result as a `.openflow.json` beside the code it explains: it opens at app.openflowkit.com
JuliusBrusseellms.txtJuliusBrussee/caveman-code/llms.txt
isolated subagents via Task tool 6. Native MCP client; caveman-code can serve as MCP server 7. Aider-grade repo map (PageRank) + edit-format-per-model 8. Daemon (`caveman serve
GarethManningllms.txtGarethManning/claude-education-skills/llms.txt
architecture, usage, and domain table - [ARCHITECTURE.md](https://github.com/GarethManning/education-agent-skills/blob/main/docs/ARCHITECTURE.md): Three-layer system design and MCP server roadmap - [EXCLUSIONS.md](https://github.com/GarethManning/education-agent-skills/blob/main/docs/EXCLUSIONS.md): What was excluded and why (learning styles, VAK, unsupported
fluffyponyPipes a download into a shellllms.txtfluffypony/dothething/llms.txt
email notifications on task completion, live input injection during execution, and a browser MCP server (dtt --browsermcp) that exposes the SearXNG + Notte stack to other agents over stdio
AVIDS2Reads credentialsllms.txtAVIDS2/memorix/llms-full.txt
only when you intentionally want repo-local guidance, rules, or hooks for one project. ### MCP server `memorix serve` starts the stdio MCP server. Use it for manual MCP wiring when
AVIDS2Reads credentialsllms.txtAVIDS2/memorix/llms.txt
memorix setup --list` lists supported agent entry points. - `memorix serve` starts the manual stdio MCP server for IDEs and agents. - `memorix background start` starts the HTTP MCP service and dashboard
CodeAbrallms.txtCodeAbra/iai-personal-memory-engine/llms.txt
query`, `episodes_recent`, `curiosity_pending`. Optional and off by default: a separate `iai-tools` MCP server (code index, session-event store, command executor), enabled with `iai-mcp capture-hooks install
unbrowse-aiReads credentialsllms.txtunbrowse-ai/unbrowse/llms.txt
would write; `setup --remove` undoes it. Restart the agent (or reload its skills and MCP servers) so the new skill and tools load. - **Sandboxed agent** (a cloud container with restricted
cryxnetReads credentialsllms.txtcryxnet/DeepMCPAgent/llms-full.txt
agentic AI systems: agents > (`build_agent()`), the tools they call (`@tool` functions and MCP servers built > with the MCP server SDK or generated from an OpenAPI spec with MCPcast
cryxnetReads credentialsllms.txtcryxnet/DeepMCPAgent/llms.txt
agentic AI systems: agents > (`build_agent()`), the tools they call (`@tool` functions and MCP servers built > with the MCP server SDK or generated from an OpenAPI spec with MCPcast
duriantacoReads credentialsllms.txtduriantaco/skylos/llms-full.txt
there are no seats or subscriptions. Credits never expire. - Interfaces: CLI, GitHub Action, MCP server, VS Code extension. ## Best repo entry points - README overview: https://github.com/duriantaco/skylos#what-is-skylos - Start
duriantacoReads credentialsllms.txtduriantaco/skylos/llms.txt
there are no seats or subscriptions. Credits never expire. - Interfaces: CLI, GitHub Action, MCP server, VS Code extension. ## Start here - Install: `pip install skylos` - Find dead code: `skylos .` - Run every
riponcmllms.txtriponcm/projectmem/llms.txt
events in plain text, feeds that history back to the agent through a native > MCP server, and warns at git commit time before you repeat an approach that > already failed
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.
