Skip to main content
Glama
robby031
by robby031

MCP Pocket Memory

Small MCP server for durable AI memory backed by SQLite.

Features

  • Store, update, delete, and search memories.

  • Optional repo field for project-specific context.

  • Optional custom tags for faster filtering.

  • Broad keyword search across key, content, repo, and tags.

  • SQLite WAL mode for light parallel agent usage.

  • Cached find_memory results with TTL.

Related MCP server: simple-memory-mcp

Build

cargo build --release

Install

From crates.io:

https://crates.io/crates/mcp-pocket-memory

cargo install mcp-pocket-memory

From GitHub Releases:

https://github.com/robby031/mcp-pocket-memory/releases

Init Database

mcp-pocket-memory init

Default database path: ~/.mcp-pocket-memory/pocket_memory.sqlite3.

MCP Config

{
  "mcpServers": {
    "pocket-memory": {
      "command": "/path/to/mcp-pocket-memory",
      "args": ["serve"],
      "env": {
        "POCKET_MEMORY_FIND_CACHE_CAPACITY": "512",
        "POCKET_MEMORY_FIND_CACHE_TTL_SECS": "30"
      }
    }
  }
}

To use a custom database path, set POCKET_MEMORY_DB.

Tools

upsert_memory

Insert or update a memory by key.

{
  "key": "repo-error-handling",
  "repo": "my-repo",
  "tags": ["convention", "error-handling"],
  "content": "Use structured errors for service boundaries. Keep user-facing messages separate from internal debug context."
}

find_memory

Find relevant memories. Can be called without filters to load recent context.

{
  "repo": "my-repo",
  "keyword": "error handling",
  "tags": ["convention"],
  "limit": 10
}

delete_memory

Delete a memory by exact key.

{
  "key": "repo-error-handling"
}

Release

GitHub Actions builds release binaries when pushing a tag:

git tag v0.4.0
git push origin v0.4.0

Maintenance

ActivitySlowing
ResponsivenessNo issues

Related MCP Connectors

Related MCP Servers

  • -
    license
    Not graded
    quality
    Not graded
    maintenance
    Provides persistent local memory functionality for AI assistants, enabling them to store, retrieve, and search contextual information across conversations with SQLite-based full-text search. All data stays private on your machine while dramatically improving context retention and personalized assistance.
    3
    -
  • A
    license
    Not graded
    quality
    C
    maintenance
    Provides AI assistants with persistent memory across sessions using local SQLite and keyword search, allowing storage and retrieval of user preferences, project context, and decisions.
    14
    8
    MIT
  • A
    license
    Not graded
    quality
    D
    maintenance
    Provides persistent memory for AI coding agents through the Model Context Protocol, enabling them to store and retrieve project knowledge across sessions.
    33
    MIT