Wiki.js MCP Server

The Wiki.js MCP Server bridges the gap between AI assistants and the powerful knowledge‑base platform Wiki.js. By exposing a Model Context Protocol (MCP) interface, it lets Claude or other Anthropic models query, create, update, and delete wiki content without ever leaving the conversational context. This is especially valuable for developers building AI‑augmented documentation tools, knowledge‑base assistants, or automated content pipelines where the model must read from and write to a live Wiki.js instance.

At its core, the server translates MCP calls into GraphQL queries against the Wiki.js API. It provides a single, uniform endpoint that understands page, user, and group operations. Developers can therefore treat Wiki.js like any other tool in the MCP ecosystem, calling functions such as or from within a conversation. The server also supports both HTTP and STDIO transports, making it suitable for web integrations as well as direct editor plugins like Cursor or VS Code.

Key capabilities include comprehensive page management—listing, searching (including unpublished), creating, updating, publishing, and force‑deleting pages—as well as user and group administration. The API exposes the full range of Wiki.js GraphQL operations, allowing fine‑grained control over content lifecycle. Because the server runs on Node.js and relies only on environment variables for configuration, it can be deployed in any containerized or serverless environment that supports Node ≥ 18.

Real‑world scenarios abound: an AI assistant can pull the latest policy documents from a corporate wiki, suggest edits, and push approved changes back; a knowledge‑base bot can surface unpublished drafts to editors for review; or a CI/CD pipeline could automatically publish documentation after successful tests. In each case, the MCP server removes friction by handling authentication, GraphQL translation, and transport management behind the scenes.

Integration is straightforward for MCP‑aware workflows. Once the server is running, any tool that implements the MCP specification can register the Wiki.js capabilities via a simple JSON configuration. The server then emits events and accepts JSON‑RPC calls, enabling seamless interaction from IDEs, chat interfaces, or custom agents. Its unique advantage lies in its dual transport support and the breadth of Wiki.js features exposed, giving developers a single point of entry for all wiki‑related AI operations.