Show HN: Renor – one memory for all your AI agents (github.com)

Renor has launched a groundbreaking tool that enables AI agents to share memory seamlessly across sessions, ensuring that what one agent learns is readily available to others. This innovative approach allows users to store decisions, corrections, and important facts in plain Markdown format on their local machines, eliminating the need for account setups or API keys. Users can easily connect Renor to multiple AI agents—including Claude Code, Codex, and Grok—within a minute using simple command-line instructions, streamlining the agent memory management process. The significance of Renor for the AI/ML community lies in its ability to enhance collaboration among various AI tools by retaining context and knowledge across different interactions. With a compact package size and storage limits, Renor also utilizes hybrid search capabilities—both keyword-based and meaning-oriented—facilitating efficient retrieval of memories. Notably, during tests with agents, Renor demonstrated impressive recall rates and accuracy, making it an invaluable resource for developers looking to create more intelligent and context-aware AI systems. Through its user-friendly design and comprehensive documentation, Renor promises to simplify the process of building memory-driven applications in AI.