# Echo Memory

Echo Memory is an MCP server for Claude Code and other agents that stores shared memory as a fact graph in Postgres. One agent's records become available to others without a model call.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 2 repo stars
- Author: echo-mem
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/echo-mem/echo-mem
- Updated: 2026-10-22
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/echo-memory

## What it does

The server stores facts as graph edges between entities: two agents who never worked together resolve names to the same entity and see what the other learned. Every fact carries metadata about who wrote it and when; writes never invoke a model. Hybrid search (vector + BM25) returns a ranked list with relevance scores and an explanation of which channels matched each fact.

Runs either via Docker with your own Postgres database or via the hosted service at `https://api.echo-mem.com/mcp` with no database to manage. Requires Python 3.11+ and environment variables `ECHO_MEMORY_DATABASE_URL`, `ECHO_MEMORY_USER_ID`, `ECHO_MEMORY_AGENT_ID`.

Seven tools: `write_episode` stores facts with no model call, `query_memory` searches by question and timestamp `as_of`, `trace_cause` builds causal chains through entities, `get_audit_log` shows change history. The `echo-memory health` command scores your store and suggests what to improve.

## Requirements
- Needs: Python and uv (commands run via uvx)
- Remote server: `https://api.echo-mem.com/mcp`
- Environment variables: `ECHO_MEMORY_DATABASE_URL`, `ECHO_MEMORY_USER_ID`, `ECHO_MEMORY_AGENT_ID` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

1. **Platform CLI**
```bash
claude mcp add echo-mem -e ECHO_MEMORY_DATABASE_URL=<ECHO_MEMORY_DATABASE_URL> -e ECHO_MEMORY_USER_ID=<ECHO_MEMORY_USER_ID> -e ECHO_MEMORY_AGENT_ID=<ECHO_MEMORY_AGENT_ID> -- uvx echo-mem
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from echo-mem <command>

2. **Platform CLI: remote server**
```bash
claude mcp add --transport http echo-memory https://api.echo-mem.com/mcp
```

### Codex

1. **Platform CLI: remote server**
```bash
codex mcp add echo-memory --url https://api.echo-mem.com/mcp
```

2. **Platform CLI**
```bash
codex mcp add echo-memory --env ECHO_MEMORY_DATABASE_URL=<ECHO_MEMORY_DATABASE_URL> --env ECHO_MEMORY_USER_ID=<ECHO_MEMORY_USER_ID> --env ECHO_MEMORY_AGENT_ID=<ECHO_MEMORY_AGENT_ID> -- uvx echo-mem
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from echo-mem <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "echo-memory": {
            "url": "https://api.echo-mem.com/mcp"
        }
    }
}
```
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "echo-memory": {
            "command": "uvx",
            "args": [
                "echo-mem"
            ],
            "env": {
                "ECHO_MEMORY_DATABASE_URL": "<ECHO_MEMORY_DATABASE_URL>",
                "ECHO_MEMORY_USER_ID": "<ECHO_MEMORY_USER_ID>",
                "ECHO_MEMORY_AGENT_ID": "<ECHO_MEMORY_AGENT_ID>"
            }
        }
    }
}
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from echo-mem <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "echo-memory": {
            "type": "http",
            "url": "https://api.echo-mem.com/mcp"
        }
    }
}
```
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "echo-memory": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "echo-mem"
            ],
            "env": {
                "ECHO_MEMORY_DATABASE_URL": "<ECHO_MEMORY_DATABASE_URL>",
                "ECHO_MEMORY_USER_ID": "<ECHO_MEMORY_USER_ID>",
                "ECHO_MEMORY_AGENT_ID": "<ECHO_MEMORY_AGENT_ID>"
            }
        }
    }
}
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from echo-mem <command>

### Gemini CLI

1. **Platform CLI: remote server**
```bash
gemini mcp add -t http echo-memory https://api.echo-mem.com/mcp
```

2. **Platform CLI**
```bash
gemini mcp add -e ECHO_MEMORY_DATABASE_URL=<ECHO_MEMORY_DATABASE_URL> -e ECHO_MEMORY_USER_ID=<ECHO_MEMORY_USER_ID> -e ECHO_MEMORY_AGENT_ID=<ECHO_MEMORY_AGENT_ID> echo-memory uvx echo-mem
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from echo-mem <command>

## When to use it
- persist facts across sessions of different agents
- search for relevant records by natural language question
- resolve entity names into a single graph without duplicates
- trace who wrote a fact and when

## Example prompts
- Record that billing uses Razorpay, not Stripe
- Show all facts about the payment module from last week
- Trace the causal chain for why performance dropped

## Similar
- [memory-lancedb-pro](https://affpapa.org/en/nevibecode/skills/memory-lancedb-pro.md) — skill · 647 installs
- [Holographic Memory](https://affpapa.org/en/nevibecode/mcp/holographic-memory.md) — MCP server · ★ 23 repo stars
- [State Memory MCP](https://affpapa.org/en/nevibecode/mcp/state-memory-mcp.md) — MCP server · ★ 44 repo stars
- [neurodivergent-memory](https://affpapa.org/en/nevibecode/mcp/neurodivergent-memory.md) — MCP server · ★ 26 repo stars
- [Memory Vault](https://affpapa.org/en/nevibecode/mcp/memory-vault.md) — MCP server · ★ 67 repo stars
- [memory-industry](https://affpapa.org/en/nevibecode/mcp/memory-industry.md) — MCP server · ★ 30 repo stars

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
