Memo

For developers

An MCP knowledge graph for your documents

Connect Claude and other MCP clients straight to your documents. Memo exposes query, ask and graph tools over the Model Context Protocol so your AI works from your real sources.

Get started freeSee pricing

Why MCP

Give your AI your documents

The Model Context Protocol (MCP) is the standard way to connect AI clients like Claude to external tools and data. Memo speaks MCP over HTTP, so any MCP-capable assistant can search your documents, ask questions across them, and read your knowledge graph — working from your real sources instead of guessing.

The tools Memo exposes

query

Semantic search across your corpus — returns the most relevant passages for a query.

ask

Question answering with citations, composed from your own documents.

list_sources

Enumerate the documents in your graph.

get_source

Fetch a single document’s metadata and chunked content.

get_graph

Export your full knowledge graph — sources, entities and relations.

Two minutes

Connect Claude to your graph

  1. Create a Personal API Token (prefix dgp_) in Settings on the Memo web app.
  2. Point your MCP client at Memo’s endpoint with the token as a bearer header.
claude mcp add --transport http memo \
  https://docgraph-api-b7e2oxlrrq-km.a.run.app/mcp \
  --header "Authorization: Bearer dgp_your_token"

Using a stdio-only client such as Claude Desktop? Bridge to the HTTP endpoint with npx mcp-remote. Full instructions are in the MCP guide.

Prefer raw HTTP?

The same graph is available over a plain HTTP API — search, ask, sources and graph endpoints, authenticated with the same token. Memo’s engine is also the open docgraph-mcp binary you can run locally.

Let Memo draw the connections

Upload your first document and watch a knowledge graph build itself. Free to start; Pro when your library grows.

Get started freeRead the docs