Skip to content
MCP server

FactoGraph

Remote MCP server for agent memory made of person-approved facts with cited answers

Install
Pricing
Free
Platforms
Web, API
Runs
Remotely
Version
0.2.0
Listing updated
1 day ago

About FactoGraph

FactoGraph is a free MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

FactoGraph stores facts only after a person has approved them, and each fact keeps its own link. When an agent asks a question, the answer cites the facts it rests on and is checked against them. If the store cannot answer, it says there is no confirmed answer rather than guessing.

The server runs no language model of its own: the connected agent does the reasoning, and the store checks new facts for contradictions with existing ones before saving. Facts can be linked in named relation layers, and derived facts record the facts they came from, so deleting a source fact also handles what depends on it.

Access works through links. A public store link can only read and ask, a private link is the owner's key to add, replace or delete facts, and a single connector URL takes the store link as an argument. The guide describes an agent flow of propose, show the person, then add, and documents of up to 150,000 characters per call. The listed price is zero.

Install

Run in a terminal

claude mcp add --transport http factograph https://api.factograph.app/api/mcp

Add to Settings → Connectors → Add custom connector

https://api.factograph.app/api/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "factograph": {
      "url": "https://api.factograph.app/api/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "factograph": {
      "type": "http",
      "url": "https://api.factograph.app/api/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "factograph": {
      "serverUrl": "https://api.factograph.app/api/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "factograph": {
      "type": "http",
      "url": "https://api.factograph.app/api/mcp"
    }
  }
}

What it does

Approved facts only
Every stored fact needs the person's approval.
Cited answers
Each answer links to the facts it relies on.
Contradiction checks
New facts are checked against stored ones.
Derived facts
Tracks facts that follow from approved ones.
Link-based access
Public links read and ask; private links can change a store.
Read-only public stores
Publishes a store others can query but not edit.

Good for

  • Give a project assistant durable decisions and deadlines
  • Let a research agent cite the exact fact behind an answer
  • Share one fact base between several agents
  • Publish approved product facts for a support agent

Pricing

Free

Listed price is $0

Prices change. Check factograph.app before you buy.

FactoGraph FAQ

Does it cost anything?

The site lists the price as 0 USD.

Do I need an account or API key?

No, the store link acts as the key.

Can an agent store facts alone?

No, each fact needs the person's approval through propose and then add.

Does it use its own language model?

No, the connected agent is the model.

Is FactoGraph free?

Yes. FactoGraph is free to use.

How do I add FactoGraph to Claude Code?

Run this command in your terminal: claude mcp add --transport http factograph https://api.factograph.app/api/mcp

What are the best alternatives to FactoGraph?

The closest alternatives to FactoGraph are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

Alternatives to FactoGraph

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  4. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  5. note1

    MCP server

    MCP server giving AI assistants search and summaries across your note1 meetings

    • 4 tools
    • Remote
    • Free trial
    • 1 GitHub star
    • Productivity