A local Mem0 MCP server for macOS with a menu bar toggle, storing memories locally in Chroma and using any OpenAI-compatible LLM for fact extraction.
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find.
Install from
M8ven verifies MCPs across every public registry — install directly from whichever one you prefer.
process.env. You'll be asked to provide them before it can run.MEM0_CHROMA_PATH— ~/.only-my-mem0ry/chroma vector store locationMEM0_COLLECTION— mem0 Chroma collection nameMEM0_DEFAULT_USER— developer_workspace default user_idMEM0_META_FILE— ~/.only-my-mem0ry/memory_meta.json sidecar: pin state + per-memory usage statsMEM0_EMBEDDER_MODEL— and cross-lingual recall. Swapping on a populated storeEVAL_VERBOSE— 1 .venv/bin/python server/eval_recall.py # per-query first-hit ranksMEM0_MCP_HOSTMEM0_MCP_PORT— 8800 MEM0_IDLE_TIMEOUT=900 ./install.shMEM0_RELATED_TOPK— 3 nearest memories add_memory surfacesMEM0_IDLE_TIMEOUT— MEM0_MCP_PORT=8800 =900 ./install.shMEM0_EMBEDDER_DIMS— MEM0_EMBEDDER_MODEL=sentence-transformers/all-MiniLM-L6-v2 =384 \MEM0_SEARCH_TOPK— 10 results search_memories returnsMEM0_DUP_THRESHOLD— , computed locally over the stored embeddings — no LLM) as primeMEM0_DUP_MAX_DOCS— 2000 skip the O(n²) duplicate scan in curate_memories above this many memoriesMEM0_HISTORY_DEPTH— capped at entries per memory (default 5; 0 disables). AMEM0_CONFLICT_LOW— ([, MEM0_DUP_THRESHOLD)) that disagree on a number, weekday,MEM0_RECENCY_BIAS— 0 opt-in recency tie-break weight over the fused ranking; 0 = off (a value <1 only breaks near-ties; measure before raising)MEM0_CONFIDENCE_BIAS— 0 opt-in confidence tie-break weight; 0 = offMEM0_HYBRID_SEARCH— with =0. No extra dependency; all local and deterministic.MEM0_RRF_K— 60 RRF constant (used only when MEM0_FUSION=rrf)MEM0_FUSION— rrf (it can reorder dense results — measure first). Turn hybrid offMEM0_BM25_MAX_DOCS— 5000 cap on lexical scan size for very large storesMEM0_CORE_FILE— ~/.only-my-mem0ry/CORE_MEMORY.md always-on core mirror file (rules files read this)MEM0_CORE_BUDGET— pin_memory(id) Pin a memory into always-on core (mirrored to a file your rules load every session). Bounded by .MEM0_MCP_TRANSPORTMEM0_SERVER_LABEL— com.only-my-mem0ry.server launchd label to start on demandMEM0_PROXY_KEEPALIVE— clamp(IDLE/3, 5, 120) seconds between keepalive pingsMEM0_BACKEND_READY_TIMEOUT— 40 seconds to wait for the backend to come upMEM0_BACKUP_KEEP[](https://m8ven.ai/mcp/ost527-only-my-mem0ry-1fhi02)