Skip to main content
Glama
unforget-ai

Unforget MCP Server

Official
by unforget-ai
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

Unforget MCP Server

Long-term memory for Claude Code, Cursor, and any MCP client. Zero LLM on write. Zero config.

Install

pipx install unforget-mcp

Related MCP server: ai-memory

Setup

Claude Code

Add to your project's .mcp.json or global settings:

{
  "mcpServers": {
    "unforget": {
      "command": "unforget-mcp",
      "args": []
    }
  }
}

Cursor

Add to Cursor settings → MCP Servers:

{
  "mcpServers": {
    "unforget": {
      "command": "unforget-mcp",
      "args": []
    }
  }
}

VS Code (Copilot)

Add to .vscode/mcp.json:

{
  "mcpServers": {
    "unforget": {
      "command": "unforget-mcp",
      "args": []
    }
  }
}

That's it. No API keys, no Docker, no database setup. An embedded PostgreSQL starts automatically.

Tools

Tool

Description

memory_store

Store a fact, preference, or decision

memory_search

Search memories with 4-channel retrieval

memory_forget

Delete memories by query or ID

memory_list

List all stored memories

memory_update

Update a memory (preserves history)

Options

# Embedded mode (default — auto-starts PostgreSQL)
unforget-mcp

# External database
unforget-mcp --database-url postgresql://user:pass@localhost/db

# Custom scoping
unforget-mcp --org-id myorg --agent-id myagent

# SSE transport (for remote connections)
unforget-mcp --transport sse --port 8767

How It Works

Claude Code / Cursor / MCP Client
    │ MCP protocol (stdio or SSE)
    ▼
unforget-mcp server
    ├── unforget core (4-channel retrieval, embedder, reranker)
    └── pgserver (embedded PostgreSQL + pgvector)

  • Memories persist in ~/.unforget/mcp-data/

  • 4-channel retrieval: semantic + BM25 + entity + temporal

  • Cross-encoder reranking for accuracy

  • Zero LLM calls on write (~7ms per memory)

License

Apache 2.0

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • ogham-mcp

    ogham-mcp
    A
    license
    Not graded
    quality
    A
    maintenance
    Persistent shared memory for AI agents. Hybrid search (pgvector + tsvector), knowledge graph, cognitive scoring, and 16-language temporal extraction. 97.2% Recall@10 on LongMemEval with one PostgreSQL query. Works across Claude Code, Cursor, Codex, OpenClaw, and any MCP client.
    22 PyPI
    115
    MIT

  • ai-memory

    alphaonedev
    A
    license
    Not graded
    quality
    A
    maintenance
    Persistent memory for any AI assistant. Zero token cost until recall. Stores memories in local SQLite, ranks by 6-factor scoring, returns results 79% smaller than JSON. Works with Claude, ChatGPT, Grok, Cursor, Windsurf, and any MCP client.
    51
    Apache 2.0

  • umo-memory

    UmoLab
    A
    license
    Not graded
    quality
    B
    maintenance
    Persistent memory for AI coding agents that stores and recalls preferences, decisions, and conventions via semantic similarity, with zero cloud dependencies and plug-and-play MCP integration for Claude Code.
    Apache 2.0