CodeAbra/iai-personal-memory-engine

Overview

A local memory engine that retains conversation history and user preferences for AI coding assistants. It stores details persistently and learns user patterns over time. Integrates with tools like Cursor, Claude Code, Codex, OpenClaw and Hermes.

Best for

Best for
Developers who use multiple AI coding assistants and want a unified, persistent memory across them.

Use cases

• Maintain consistent context across multiple AI coding sessions
• Let an AI assistant recall user’s preferred coding style and conventions
• Build a persistent long-term memory for local AI agents

How to use

Install

pip install .

Tools exposed

• IAI_MCP_STORE
• IAI_MCP_PYTHON
• IAI_MCP_RECALL_CONCURRENCY
• IAI_MCP_RECALL_SLOT_WAIT_SEC

Tested with

Claude Desktop, Claude Code, Cursor, Cline, Continue

Notes

A local memory engine that retains conversation history and user preferences for AI coding assistants. It stores details persistently and learns user patterns over time. Integrates with tools like Cursor, Claude Code, Codex, OpenClaw and Hermes.

340 stars on GitHub. Last updated 2026-07-13. Licensed MIT.

Use cases

• Maintain consistent context across multiple AI coding sessions
• Let an AI assistant recall user’s preferred coding style and conventions
• Build a persistent long-term memory for local AI agents

Pros

• Free and open source under MIT license
• Runs locally, keeping data private and offline
• Compatible with several popular AI development tools

Cons

• Requires Python environment and local setup
• May lack support for less common AI assistants or custom workflows
• Memory reliability depends on local storage uptime and maintenance

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.