Skip to content
MCP server

Echo Memory

Shared graph memory for AI agents, stored in your own Postgres

Pricing
Freemium
Platforms
self-hosted, CLI, API
Runs
Locally or Remotely
Version
0.6.1
Listing updated
3 days ago

About Echo Memory

Echo Memory is a freemium MCP server for knowledge and memory. It can run locally on your machine or as a hosted remote server.

Echo Memory models what an agent learns as a graph: entities are nodes and each fact is an edge between two of them. Names resolve to a single node per scope, so a session that never met an earlier one still inherits what it learned. Facts record which tool wrote them and when, and a superseded fact is kept with its history rather than deleted.

The design choice that sets it apart is that writing never calls a model. Extraction happens in the calling agent, which has to arrive with entities and facts already structured; the site states this trade-off openly. Retrieval uses vector similarity and graph traversal in one Postgres database, and the author reports on their own store that queries returned 96.7% less context with the answer still present 87.2% of the time.

It needs Python 3.11 and Docker for a local database, started by an echo-memory quickstart command, or you can use the hosted service. Seven tools, including write_episode and query_memory, let Claude Code, Cursor and Codex share one memory.

Install

Run in a terminal

claude mcp add --transport http echo-mem https://api.echo-mem.com/mcp

Add to Settings → Connectors → Add custom connector

https://api.echo-mem.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

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

Add to .vscode/mcp.json

{
  "servers": {
    "echo-mem": {
      "type": "http",
      "url": "https://api.echo-mem.com/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "echo-mem": {
      "serverUrl": "https://api.echo-mem.com/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "echo-mem": {
      "type": "http",
      "url": "https://api.echo-mem.com/mcp"
    }
  }
}

Configuration

  • ECHO_MEMORY_DATABASE_URLPostgres connection string. `echo-memory quickstart` starts a suitable database and prints this.Required
  • ECHO_MEMORY_USER_IDWhose memory this is. Scopes every read and write.Required
  • ECHO_MEMORY_AGENT_IDWhich tool is writing: claude-code, cursor, codex, claude-desktop. Recorded on every fact, and what makes a cross-tool recall visible afterwards.Required

Tools this server provides (2)

Taken from the project documentation; the running server may differ.

write_episode
Records entities and facts as graph nodes and edges.
query_memory
Retrieves the connected facts relevant to a question.

What it does

Graph data model
Entities are nodes and facts are edges, so retrieval is bounded by what connects to the question.
No model on write
Storing a memory invokes no LLM; the calling agent supplies the extracted facts.
Authorship tracking
Each fact records writer, tool, project and time.
Superseded facts kept
Replaced facts stop being returned but stay reachable with history.
One database
Vector similarity and graph traversal run in your own Postgres.
Cross-tool sharing
Different agents write to and read from the same graph.

Good for

  • Share project knowledge between Claude Code, Cursor and Codex
  • Give a support bot memory of what a customer was told
  • Record why an ops step exists for later incidents
  • Keep decisions that change over time with their history

Pricing

Freemium, from $149 a month

Free open source self-hosting; hosted service $149 a month

Prices change. Check echo-mem.com before you buy.

Echo Memory FAQ

Where is the data kept?

In your own Postgres database when self-hosted, or run for you on the hosted service.

What does it cost?

The hosted service is $149 a month; self-hosting is open source and free at any size.

What do I need to run it locally?

Python 3.11 or newer and Docker for the database.

Why is the PyPI package named echo-mem?

The README says the name echo-memory belongs to an unrelated product, so the distribution is echo-mem.

Is Echo Memory free?

Echo Memory has a free plan, with paid plans for more usage or features. Paid plans start at $149 a month.

How do I add Echo Memory to Claude Code?

Run this command in your terminal: claude mcp add --transport http echo-mem https://api.echo-mem.com/mcp

What are the best alternatives to Echo Memory?

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

Alternatives to Echo Memory

  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. Midplane

    MCP server

    Postgres gateway for AI agents with table grants, column masking and approval for writes

    Runs locally
    • Open source
    • 5 GitHub stars
    • Databases
  6. note1

    MCP server

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

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