Local-first long-term memory database for AI systems, providing SQLite FTS5/BM25 retrieval and optional SentenceTransformer embeddings via MCP server.
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.CODEX_SANDBOXMINIFLARE_REGISTRY_PATHWRANGLER_LOG_PATHWRANGLER_WRITE_LOGSOPENAI_BASE_URLOPENAI_API_KEY— First-party OpenAI API; the endpoint is fixed and is runtime-only.NO_PROXYno_proxyNARRATORDB_DATA_DIR— Data directory; defaults to ~/.narratordbNARRATORDB_PATH— Explicit database fileNARRATORDB_DB_PATH— Equivalent explicit database fileNARRATORDB_USER_ID— Default logical user/scopePYTHONDONTWRITEBYTECODENARRATORDB_COMPILER_MAX_COST_USD— Optional per-process cumulative soft USD compiler stop; unset by default, and an in-flight call can finishNARRATORDB_EMBEDDING_MODEL_DIR— Explicit local embedding model directoryNARRATORDB_LOCAL_ONLY— Require locally cached embedding assets when trueNARRATORDB_EMBEDDING_MODEL— SentenceTransformer model identifierNARRATORDB_AUTO_CAPTURE— behavior. =false disables every automatic write whileNARRATORDB_DEBUGNARRATORDB_MCP_CLIENTNARRATORDB_WORKSPACE_ID— set when that scope must be portable.NARRATORDB_ALLOW_PATH_FALLBACK_WRITES— true before starting Codex. A non-home,[](https://m8ven.ai/mcp/williamjnsson-narratordb-ubmez7)