imprint-memory is an open-source memory layer for LLM agents, capturing every conversation turn and supporting multimodal embedding (text and image). It offers hybrid search capabilities and an auto-surfacing recall hook, with both stdio and HTTP MCP server support. Designed for AI developers building agent frameworks.
imprint-memory is an Other AI product. It focuses on enabling LLM agents to automatically capture, embed, and recall conversation history for improved context and memory. imprint-memory is an open-source project aimed at AI developers and agent framework builders. The project is open source (AGPL-3.0-or-later). imprint-memory is available on the command line and API, and it can be self-hosted.
Behind imprint-memory is Qizhan7, and the product first shipped in 2026. The project is developed in the open on GitHub with 58 stars and 98 commits in the last 90 days. Among its 5 catalogued features are conversation capture, multimodal embedding, and hybrid search. It exposes integrations via an MCP server and a public API.
Latest indexed changes and source events
imprint-memory verified by the PulseGate indexer
Other apps tracked under the same category.
