Show HN: Renor – one memory for all your AI agents

Category: ai-ml

Tags: mcp, ai-agents, memory, markdown, search

Score: 6.8/10 (Innovation: 6, Technical: 6, Documentation: 8, Utility: 7)

Renor is a shared memory layer for coding agents, implemented as an MCP server that stores decisions and facts as plain Markdown and makes them searchable across sessions. It stands out for combining a simple, self-hosted architecture (SQLite keyword index, no embedding model required) with strong hybrid search benchmarks and broad client support (Claude Code, Cursor, Codex, etc.).

Target audience: developers using AI coding assistants, AI engineers, devops

Repository: https://github.com/renorai/renor-engine · TypeScript · Apache-2.0 · 26 stars

View on Hacker News