Open-source, local-first memory and context for AI agents with scoped memory, provenance, and MCP/HTTP access.
remnic is worth checking the docs before setup with trust notes worth reviewing. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view joshuaswarren/remnic --webOpen the official README and confirm the supported install method.
Add the server entry to your MCP client config.
Restart your agent and verify that the server tools appear.
Repository setup guidance
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 24 days ago.
168 GitHub stars indicate community interest.
70 open issues signal maintenance load.
MIT license detected.
3 security/trust notes recorded.
Setup difficulty is 3/5.
Codebase intelligence layer for AI coding agents: code health, git analytics, dead code detection, and architectural decisions via MCP.
Deep code indexing MCP server for AI agents with hybrid FTS5 + embedding search, call graphs, and multi-repo workspaces.
A project template that turns your repository into stable infrastructure for AI-assisted development.
Local code intelligence MCP server and CLI for AI coding agents, providing semantic search and call graph analysis.
Remnic gives AI agents a shared memory that lives on your own computer as plain markdown files. It watches conversations, saves important information, and brings it back when needed. You can use it with many AI tools, and you control what is stored and shared.
Remnic is an open-source, local-first memory and context system designed for AI agents. It provides a unified memory store that works across multiple AI tools and agents, including OpenClaw, Claude Code, Codex CLI, Cursor, ChatGPT (developer mode), Hermes, Replit, and any MCP client. Every memory is stored as a plain markdown file with YAML frontmatter on your local disk, ensuring full ownership, transparency, and version control without any cloud dependency or subscription. The system features automatic extraction and recall, watching conversations to distill durable knowledge and inject relevant context when needed. It offers sharp retrieval through hybrid search combining BM25, vector search, and reranking over rebuildable indexes, along with graph recall, memory-worth scoring, and per-result provenance for inspection. Remnic includes scoped memory, provenance tracking, retrieval quality controls, correction mechanisms, boundaries, evals, and MCP/HTTP access. It is MIT licensed and designed to be forked and extended. The project also includes a 'What Helps Me' feature, a private support passport that lets users share selected memories as short first-person cards with strict consent and revocation controls.
168
Stars
23
Forks
70
Issues
MIT
License
Open-source, local-first memory and context for AI agents with scoped memory, provenance, and MCP/HTTP access.
Developers building AI agents or assistants, Power users of AI coding tools like Claude Code or Codex, Teams integrating MCP servers into their workflows
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
Stores conversation data locally; ensure proper access controls on your machine Automatic extraction may capture sensitive information; review and curate memories
Similar or complementary options to evaluate include repowise, srclight, attractor-guided-engineering-template.
Aug 20, 2026