Repository navigation
Using claude-mem worker API for multi-agent memory enrichment #1558
ThinkOffApp
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Built on OpenClaw. We are using the claude-mem worker API as the memory backend for ide-agent-kit.
The integration works like this: when a message arrives in a GroupMind room, the enrichment layer calls
GET /api/search/observationswith the message body as the query. The top results get injected into the event payload before the IDE agent (Claude Code, Codex, Cursor, Gemini) processes it.The result is that agents have relevant context from previous sessions when responding to team messages. Combined with our user-intent-kit integration (cross-device intent state), agents adapt both what they know and how they respond.
Config is two lines in the IAK config file:
memory_api.baseUrl: your claude-mem worker URLmemory_api.token: auth tokenWould be interested in hearing about other integrations people have built with the worker API.
All reactions