Vibe Coding Discoverremnic
View on GitHubOpen-source memory and context for user-aware agents: scoped memory, provenance, retrieval quality, correction, boundaries, evals, and MCP/HTTP access.
Remnic provides local-first, Markdown-backed memory for AI agents, with hybrid search, provenance, correction and retrieval-quality controls. Agents access the shared store through MCP or HTTP integrations.
Use Cases
Share persistent user context across AI agentsRetrieve relevant memories across conversationsCorrect stale or conflicting agent memoriesInspect the provenance of recalled factsBuild a local-first memory store for agent workflows
Built With
- Language
- TypeScript
- Frameworks
- MCP · OpenClaw · Claude Code · Codex CLI · Cursor
Tags
agent memory · long-term memory · semantic search · hybrid retrieval · knowledge graph · provenance · local-first · memory correction · MCP server · Markdown