mnemo
@jmeiracorbal
About mnemo
Persistent memory for AI coding agents. mnemo stores decisions, bugs, conventions, and discoveries across sessions in a local SQLite database.
Basic information
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"mnemo": {
"command": "mnemo",
"args": [
"mcp",
"--tools=agent"
]
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is mnemo?
mnemo is an MCP server that provides persistent memory for AI coding agents by storing decisions, bugs, conventions, and discoveries across sessions in a local SQLite database. It integrates with Claude Code and Cursor via hooks and MCP, giving agents a continuous memory that survives across conversations.
How to use mnemo?
Install mnemo using the one-line installer (curl -sSf .../install.sh | bash), which downloads the binary, installs it to ~/.local/bin/, and runs mnemo setup automatically. For Claude Code run mnemo setup; for Cursor run mnemo setup --cursor. After setup, restart the editor and mnemo's MCP tools become available under the mcp__mnemo__* namespace.
Key features of mnemo
- Session hooks automatically start/end sessions and inject memory context at the beginning of every conversation
- 14 MCP tools:
mem_save,mem_search,mem_context,mem_session_summary, and more - Passive capture extracts learnings from conversation transcripts automatically at session end
- Full CLI: save, search, export, import, and inspect memories from the terminal
- Own isolated storage at
~/.mnemo/memory.dbcreated automatically on first run - Native integration for both Claude Code and Cursor via their respective hook systems
Use cases of mnemo
- Save a coding decision (e.g., "Use FTS5 for search") with a project tag for later retrieval
- Search all past memories with full-text search to quickly find relevant context in a new session
- Automatically capture learnings from subagent output without manual saving
- Export all memories to JSON for backup or migration between machines
FAQ from mnemo
What runtime dependencies does mnemo require?
Go 1.22+ is required to build from source, but the prebuilt binary has no runtime dependencies beyond macOS or Linux. For editor integration, Claude Code CLI or Cursor 2.6+ must be installed.
How does mnemo store memories?
mnemo uses a local SQLite database with FTS5 full-text search, stored at ~/.mnemo/memory.db. The directory and database are created automatically on first run with no manual setup required.
How do I install mnemo?
You can install with the one-line installer (curl -sSf .../install.sh | bash), via the Claude Code plugin marketplace (claude plugin marketplace add jmeiracorbal/mnemo), or by building manually from source with go build.
Can mnemo be used with Cursor?
Yes, run mnemo setup --cursor to install hooks and configure the MCP server in Cursor 2.6+. You can preview changes with mnemo setup --cursor --dry-run.
What does passive capture do?
At session end, mnemo reads the conversation transcript (via the stop hook) and automatically extracts and saves learnings from the text, without requiring any explicit save command from the user.
More Databases MCP servers
Redis MCP Server
redisThe official Redis MCP Server is a natural language interface designed for agentic applications to manage and search data in Redis efficiently
MCP Alchemy
runekaagaardA MCP (model context protocol) server that gives the LLM access to and knowledge about relational databases like SQLite, Postgresql, MySQL & MariaDB, Oracle, and MS-SQL.
Meilisearch MCP Server
meilisearchA Model Context Protocol (MCP) server for interacting with Meilisearch through LLM interfaces.
MySQL MCP Server
designcomputerA Model Context Protocol (MCP) server that enables secure interaction with MySQL databases
mcp-server-qdrant: A Qdrant MCP server
qdrantAn official Qdrant Model Context Protocol (MCP) server implementation
Comments