MarcoPorcellato/matryca-plumber — Local-first AI daemon for Logseq OG: background semantic indexing, link hygiene, and agent-ready CLI…

Description and use

**Brief Summary**
Matryca‑Plumber is a local‑first AI daemon that enriches a Logseq knowledge base by performing background semantic indexing, link‑hygiene checks, and exposing a Model Context Protocol (MCP)‑compatible CLI/SDK. Written in Python, it edits Markdown files directly on disk—no cloud services or Logseq API are required—making it a lightweight bridge that lets LLM agents act on real user data and tools.

**Value**

• **Standardised AI‑tool integration** – By implementing the open Model Context Protocol, Matryca‑Plumber gives developers a common, language‑agnostic interface for connecting LLM agents to local files, scripts, or external services.
• **Local‑first privacy & performance** – All indexing and edits happen on the user’s machine, avoiding data‑exfiltration risks and latency associated with cloud‑based pipelines.
• **Ready‑to‑use CLI/SDK** – The daemon can be started as a background service and queried from any language that can call its HTTP/CLI endpoints, simplifying the creation of custom agents, bots, or automation workflows.

**Practical Adoption Path**

1. **Prototype** – Clone the repo, install the Python dependencies, and run the daemon locally against a Logseq vault. Use the provided CLI (matryca) to issue simple queries (e.g., “find all notes about X”).
2. **Integrate** – Wrap the CLI or call the exposed MCP HTTP API from your AI agent or automation script. Because the protocol is language‑agnostic, you can embed it in Node, Go, or Rust services without rewriting the core daemon.
3. **Scale** – Deploy the daemon as a systemd service or Docker container on a workstation or edge server, configure it to watch multiple Markdown roots, and expose the MCP endpoint to internal tools (e.g., CI pipelines, internal chat‑ops bots).
4. **Extend** – Contribute additional indexing plugins or link‑hygiene rules via the documented plugin hooks, or replace the default LLM backend with a custom model for tighter security or cost control.

**Production Readiness**

• **Activity & Community** – The project shows recent commits (last updated 2026‑06‑22), 67 stars, and a modest but active fork base, indicating ongoing maintenance.
• **Maturity** – Core features (semantic indexing, link hygiene, MCP server) are already implemented and documented; the Python codebase is concise and leverages well‑known libraries, reducing the surface area for bugs.
• **Risk Profile** – No immediate licensing or security red flags appear, though a formal audit of the daemon’s network surface and dependency chain is advisable before enterprise rollout.
• **Readiness Verdict** – With its clear API, local‑first design, and active upstream, Matryca‑Plumber is a strong OSS candidate for pilots and can be promoted to production after a brief security review and integration testing.

Problem

MarcoPorcellato/matryca-plumber helps connect AI assistants to real tools and data through a standard protocol.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Evidence

Local-first AI daemon for Logseq OG: background semantic indexing, link hygiene, and agent-ready CLI/MCP — edits Markdown on disk (no cloud, no Logseq API). Karpathy LLM-Wiki inspired.

Source