Loop Memory is a local memory system for AI agents that captures conversations, scores memories using importance, recency, usage, and feedback, and distils them into a curated wiki. It provides a web UI and supports agent workflows including Codex, Claude, Hermes, and OpenClaw.
In the Memory & skills space, Loop Memory takes a focused approach. It focuses on maintaining persistent, searchable long-term memory across AI agent conversations. It is built as an open-source project for AI agent developers. Loop Memory is open source under the MIT license. Loop Memory is available on the command line and the web, and it can be self-hosted.
It is developed by smartfind, and it first shipped in 2026. The project is developed in the open on GitHub with 114 commits in the last 90 days. Among its 8 catalogued features are conversation capture, importance scoring, and recency scoring.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do