iai-mcp
Active·★ 133·MIT·Updated 2026-05-15
★ Trending★ Code Assistant★ RAG / Knowledge Base
The best-benchmarked open-source memory system for AI coding assistants
IAI-MCP is a local memory layer for MCP-compatible AI assistants. It captures every conversation turn verbatim, organizes them over time, and serves relevant memories at the start of new sessions. All data is encrypted locally with AES-256-GCM and never leaves your machine.
#ai-agents#ai-memory#claude#claude-code#embeddings#episodic-memory#lancedb#llm-tools
01
Features
01Ambient capture of all conversation turns with timestamps and metadata
02Automatic recall of relevant memories at session start
03Fully local, no telemetry or cloud dependency
04Encrypted at rest with AES-256-GCM
05Verbatim recall >=99% at 10k episodes
02
Compatibility
Claude Code
Claude Code
Verified via docs
Claude Desktop
Claude Desktop
Verified via docs
Codex CLI
Codex CLI
Verified via docs
Other MCP-over-stdio hosts
Other MCP hosts
Verified via docs
03
Quick start
1
$ git clone https://github.com/CodeAbra/iai-mcp.git
2
$ cd iai-mcp
3
$ bash scripts/install.sh
04
Use cases
↳Provide long-term memory for Claude Code, Codex CLI, and other MCP assistants
↳Eliminate the need to repeat context across sessions
↳Personalize AI assistant behavior over time based on past interactions
05
Alternatives
Brave Search MCP★ 86.5k
Allow your AI Agent to search the real-time internet using Brave Search API. Essential for getting up-to-date information.
context-mode★ 16.0k
Context window optimization for AI coding agents. Sandboxes tool output, 98% reduction. 12 platforms
Auto-claude-code-research-in-sleep★ 11.0k
ARIS ⚔️ (Auto-Research-In-Sleep) — Claude Code skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation via Codex MCP
agents-best-practices★ 1.1k
Provider-neutral Agent Skill for Codex, Claude Code, and agentic harness design.
Related searches
Comments
Log in to leave a comment
- SSam ClarkMay 25, 2026
Best-benchmarked open-source memory system for AI coding assistants — strong claim.
- LLogan DavisMar 21, 2026
Works with AI coding assistants as a drop-in memory layer.
- PPeyton ClarkMar 6, 2026
Good for developers who want memory that's been rigorously evaluated, not just built.
- SShawn JacksonFeb 26, 2026
Benchmark-driven development means performance is measured, not assumed.