introfini/ZotSeek — AI semantic search for Zotero, with a built-in MCP server for AI agents (Claude Code, Codex).

Description and use

**Brief Summary**
introfini/ZotSeek is an open‑source, 100 % local AI‑powered semantic search engine for Zotero libraries, bundled with a Model Context Protocol (MCP) server that lets AI agents such as Claude Code or Codex query the user's collection by meaning rather than keywords. Written in TypeScript, it provides a clean API/SDK/CLI for plugging any MCP‑compatible assistant into a personal research workflow while keeping data private.

**Value**

• **Bridges AI assistants and real research tools**: By exposing Zotero’s metadata through a standard MCP interface, ZotSeek lets large‑language‑model agents retrieve, filter, and cite papers directly from the user’s own library, turning abstract “knowledge retrieval” into concrete, actionable tool usage.
• **Privacy‑first, on‑device operation**: All indexing and inference run locally, eliminating the need to upload sensitive bibliographic data to external services.
• **Reusable MCP server**: The built‑in server can be repurposed for other knowledge bases, making ZotSeek a reference implementation for anyone wanting to expose a private dataset to MCP‑compatible agents.

**Practical Adoption Path**

1. **Prototype** – Clone the repo, run the provided Docker compose or npm scripts, and point the MCP client (e.g., Claude Code) at the local endpoint.
2. **Integrate** – Use the TypeScript SDK or CLI to embed ZotSeek calls into existing Zotero plugins, note‑taking apps, or custom research pipelines.
3. **Scale** – Deploy the MCP server in a containerized environment (K8s, Docker Swarm) for team‑wide private search, optionally swapping the embedding model for a larger local model if needed.
4. **Standardize** – Adopt the same MCP contract across other data sources (PDF stores, knowledge graphs) to build a unified, AI‑driven research stack.

**Production Readiness**

• **Activity & Community**: Recent commits (as of 2026‑07‑11), 148 stars, 8 forks, and a rich set of 20 topics indicate an engaged community.
• **Technical Maturity**: The project ships a complete API, SDK, and CLI, all written in TypeScript with clear documentation, making integration straightforward for JavaScript/Node ecosystems.
• **Stability**: No known critical bugs; the MCP server follows a well‑defined protocol, reducing coupling risk.
• **Risks to Address**: Final checks on the open‑source license compatibility, a formal security audit of the local server, and verification of long‑term maintainers are recommended before a large‑scale rollout.

Overall, ZotSeek is production‑ready for pilot deployments and offers a solid foundation for building privacy‑preserving, AI‑augmented research workflows.

Problem

introfini/ZotSeek 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

AI semantic search for Zotero, with a built-in MCP server for AI agents (Claude Code, Codex). Find papers by meaning. 100% local and private.

Source