An MCP server that stores a tree-structured knowledge base as an editable book of sections and content nodes. It supports browsing, creating, updating, moving, querying, exporting/importing, and versioning nodes, with selected node bodies automatically injected into session context via the `inject=true` property.
From the registry: Tree-structured knowledge base as an MCP server
$ cargo install --path .Please install the `outline-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Rust and Cargo** — Required to build and install the native outline-mcp binary via cargo. (https://www.rust-lang.org/tools/install/)
- **Install outline-mcp binary** — Build and install the server locally using Cargo from the project source. Run: `cargo install --path .`
Optional prerequisites:
- Optional Docker runtime — Docker can be used instead of installing Rust and building the binary locally. (https://www.docker.com/)
Canonical MCP server config (stdio transport):
- command: `outline-mcp`
- args: ["<path-to-your-book.json>"]
Note: Supports stdio transport. The simplest native setup is the installed binary: command `outline-mcp` with a single optional path argument to the book JSON file. If the path argument is omitted, it defaults to `outline-book.json` in the current working directory. The README also documents a Docker-based stdio launch alternative using `docker run -i --rm -v <path-to-data>:/data ghcr.io/ynishi/outline-mcp:latest /data/<your-book.json>`, but the native binary is the simpler canonical shape. No environment variables are documented.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.MCP server for searching Airweave collections with natural language queries.