MARM
Local-first persistent memory layer for AI agents
Visit lyellr88.github.io ↗What is MARM?
MARM is a memory system that gives AI agents persistent storage and recall capabilities without cloud dependency. It works as an MCP server compatible with Claude, Codex, Gemini, Grok, and other agents, storing session history, codebase indexes, and concept graphs in SQLite. The system runs locally for privacy and instant access, with support for solo developers, multi-agent swarms, and high-throughput scenarios.
Setup takes about 60 seconds via pip install and initialization commands. MARM includes a web console for viewing memory state, 16 MCP tools for agent interaction, and knowledge graph features for organizing concepts. The project welcomes contributions and maintains active documentation with troubleshooting guides and architecture internals.
Momentum
Daily activity over the last 30 days: upvotes, unique visits, new GitHub stars and Hacker News points.
How do we know it's vibe coded?
vibetrend discovered MARM automatically. Evidence: GitHub Lyellr88/marm-memory; repo has AGENTS.md. View source.
Similar vibe-coded apps
Turn your codebase into a queryable knowledge graph for AI coding assistants
Operating layer for AI coding agents with skills, memory, planning and security
Self-hosted AI agent with persistent memory and multi-platform messaging
Open-source AI platform for short-form video and animated series creation
Open-source canvas workbench for AI-powered image and video creation
Community marketplace for DeepSeek Harness Web themes, pets, plugins
Open-source, AI-native knowledge base with Evernote-style interface