Searches, reads, and writes long-term character memories in a Qdrant vector database.
npx skills add https://github.com/momori777/Artemis --skill mem0-bridge-momori777
Please help me install this Agent Skill. Skill: mem0-bridge Source: https://github.com/momori777/Artemis/tree/main/skills/mem0-bridge Command: npx skills add https://github.com/momori777/Artemis --skill mem0-bridge-momori777
High-intent search queries and answers about installing and using this skill.
Use the add_memory function with a character user_id and fact string; the text is embedded by a local all-MiniLM-L6-v2 server and written to Qdrant. Search later with search_mem0_qdrant to retrieve relevance-scored memories for prompt injection.
Each character maps to a distinct user_id (sakura, natsume, enola, atori) in the Qdrant collection, so reads and writes are scoped per character. This prevents memory leakage between different AI personas sharing one database.
Search returns results with all scores at 0.0 when the embedding server on port 9999 is down. The integration module detects this case and injects the top results as low-relevance fallback references instead of returning empty context.
Empty context occurs when no memories exist for that character's user_id or the relationship.json behavior state file is missing. Verify Qdrant data exists under the memory directory and the embedding server is running on port 9999.
The bridge requires the qdrant-client package installed via pip and a running local embedding server on port 9999 serving all-MiniLM-L6-v2 with 384-dimensional vectors. No cloud API keys are needed since everything runs locally.
SillyHippyIndexes legal case data into Qdrant and retrieves similar cases via semantic search.
bbvch-aiDocuments Swiss AI Hub bot framework architecture, handlers, and multi-channel messaging logic.
ste-bahManage Archon behavioral rule priorities, tiers, conflicts, and lifecycle.
winsorllcIndex documents with embeddings and search them by semantic similarity.
