About kj455/mcp-kibela
mcp-kibela is a Model Context Protocol (MCP) server implementation that enables AI assistants like Claude to search and reference content stored in Kibela, a knowledge management platform. It provides a set of tools for interacting with Kibela notes, including searching by keywords, fetching the current user's latest notes, retrieving note content and comments by ID, accessing notes by path, creating new notes, and updating existing note content. The server can be deployed via Docker or run directly with Node.js, and requires a Kibela team name and API token for authentication. It is designed for integration with MCP-compatible clients such as Claude Desktop, Cursor, and VS Code.
Key Features
Search Kibela notes by keywords
Fetch the current user's latest notes
Retrieve note content and comments by note ID
Get note content by path
Create new notes in Kibela
Update existing note content by note ID
Docker container deployment for easy setup
Compatible with Claude Desktop, Cursor, VS Code, and other MCP clients
Open source under MIT license
Pros & Cons
Pros
- Supports a wide range of note operations (search, fetch, create, update)
- Easy deployment via Docker or directly with Node.js
- Works with popular MCP-compatible clients out of the box
- Open source with MIT license for customization and community contributions
- Clear configuration with environment variables (team name and API token)
Cons
- Requires a Kibela account and API token, limiting use to Kibela users
- Only supports Kibela (not generic knowledge base integration)
- Requires Node.js v18 or higher if running without Docker
- Limited documentation beyond README (no dedicated website or examples)
Best For
Integrate Kibela knowledge base with AI coding assistants like Claude DesktopAutomate note creation and updates from AI agents or chat interfacesSearch company documentation via natural language queries in Cursor or VS CodeProvide AI with secure access to internal Kibela notes for context-aware responses
FAQ
What is mcp-kibela?
mcp-kibela is an MCP (Model Context Protocol) server that allows AI assistants to search, read, create, and update notes in Kibela, a knowledge management platform.
How do I install and configure mcp-kibela?
You can run it via Docker: `docker run -i --rm -e KIBELA_TEAM -e KIBELA_TOKEN ghcr.io/kj455/mcp-kibela:latest`. Or clone the repo and run with Node.js. Set environment variables KIBELA_TEAM (your team name from https://[team].kibe.la) and KIBELA_TOKEN (your Kibela API token).
Which AI clients are compatible?
mcp-kibela works with any MCP-compatible client, including Claude Desktop, Cursor, and VS Code. Configuration examples for each are provided in the README.
What features does mcp-kibela offer?
It provides note search by keywords, fetching your latest notes, getting note content and comments by ID, accessing notes by path, creating new notes, and updating note content by ID.