Hemory is a listening and memory platform designed to turn conversations and everyday experiences into searchable source material for AI agents. Its app captures audio in manual or scheduled listening modes, then organizes what it hears into private memories that users can search and use to generate documents, reports, journal entries, product briefs, and other content. Hemory connects to compatible agents through the Model Context Protocol (MCP), including Claude Code, Codex, Gemini, Cursor, VS Code, OpenClaw, and Hermes, allowing users to ask questions about past conversations or create work grounded in real events.

Privacy is central to the service: audio is processed as a stream and is not retained in the cloud, while raw recordings remain on the listening device and are not synced across devices. Hemory offers a free trial with 10 hours of listening, plus paid Starter, Pro, and Max plans with increasing monthly allowances. Voice-activity detection counts speaking time rather than silence. iOS and Android apps are available; additional platforms and self-hosting are listed as coming soon.