agentleFS
IBazylchukIBazylchuk/paparats-mcp/llms.txt

paparats-mcp

llms.txt34 starsChanged 19 days agoMITView on GitHub

Paparats MCP is an open-source (MIT), self-hosted MCP server for semantic code search across many repositories, for Claude Code, Cursor, Codex, GitHub Copilot and any other MCP client. It indexes code with tree-sitter AST chunking, builds a cross-chunk symbol graph, attaches git history and ticket references to every chunk, and stores vectors in Qdrant using local llama.cpp embeddings. Code never leaves the machine by default. There is no hosted version.

  • Installs packages

The file

We split this llms.txt into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Not included: code editing or refactoring tools, BM25 keyword search on the code layer (the docs layer is hybrid)

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 3,806 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
llms.txt
# Paparats MCP

> Paparats MCP is an open-source (MIT), self-hosted MCP server for semantic code search
> across many repositories, for Claude Code, Cursor, Codex, GitHub Copilot and any other
> MCP client. It indexes code with tree-sitter AST chunking, builds a cross-chunk symbol
> graph, attaches git history and ticket references to every chunk, and stores vectors in
> Qdrant using local llama.cpp embeddings. Code never leaves the machine by default.
> There is no hosted version.

## Docs

- [README](https://github.com/IBazylchuk/paparats-mcp#readme): what it is, quick start, what the agent gets
- [Install](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/install.md): install options, plain docker compose, Codespaces, support mode, hardware
- [Clients](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/clients.md): setup for Claude Code, Cursor, Codex, Copilot, Windsurf, Antigravity, Gemini CLI and others
- [Configuration](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/configuration.md): projects.yml, .paparats.yml, groups, freshness, CLI, environment variables
- [How it works](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/how-it-works.md): indexing, search and ranking, the call graph, the module map
- [Agent install guide](https://github.com/IBazylchuk/paparats-mcp/blob/main/llms-install.md): step-by-step install for an AI agent
- [MCP tools](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/tools.md): tools per endpoint
- [FAQ](https://github.com/IBazylchuk/paparats-mcp#faq): privacy, freshness, which branch is indexed, grep, hardware, supported clients
- [Comparison with alternatives](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/comparison.md): Graphify, codegraph, GitNexus, codebase-memory-mcp, claude-context, Serena, ast-grep, Sourcegraph
- [Chunking strategy](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/chunking-strategy.md)
- [Replacing embedding models](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/replacing-embedding-models.md)
- [Monitoring and observability](https://github.com/IBazylchuk/paparats-mcp/blob/main/docs/observability.md)
- [Privacy](https://github.com/IBazylchuk/paparats-mcp/blob/main/PRIVACY.md)

## Facts

- Install: download examples/docker-compose/docker-compose.yml and projects.yml, then `docker compose up -d` (needs only Docker with Compose v2); or the CLI: `npm install -g @paparats/cli && paparats install`, then `paparats add <path-or-github-repo>` (needs Node.js 24+)
- Package: @paparats/cli (npm); MCP Registry id io.github.IBazylchuk/paparats-mcp; license MIT
- Endpoints (Streamable HTTP, default port 9876): /mcp (coding), /support/mcp (support), /knowledge/mcp (docs-only, opt-in, bearer auth)
- Embeddings: bge-code-v1 for code (1536d), Qwen3-Embedding-0.6B for docs, glossary and architectural memory (1024d), served locally by llama.cpp llama-server + llama-swap; OpenAI and Voyage are opt-in
- Languages with AST chunking and symbol extraction: TypeScript (incl. TSX), JavaScript, Python, Go, Rust, Java, Ruby, C, C++, C#, Terraform/HCL
- Freshness: the indexer checks each repo for changes every 10 minutes (git ls-remote or file mtime/size fingerprint) and runs a full pass every few hours; POST /trigger on port 9877 reindexes on demand; there is no file watcher
- Multi-repo: projects in the same group share one Qdrant collection and are searched together
- Not included: code editing or refactoring tools, BM25 keyword search on the code layer (the docs layer is hybrid)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

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.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.