Marrow is a persistent multi-project memory backend that provides AI agents with MCP tools for semantic code navigation, task backlog management, and versioned artifact storage. It enables seamless, token-optimized context sharing and asynchronous handoffs between different AI models working on the same codebase.
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.GITHUB_TOKENSECRET_TOKEN— your-strong-random-secretEMBEDDING_MODEL_CODE— Sentence-transformer model for code skeleton embeddings BAAI/bge-small-en-v1.5EMBEDDING_MODEL_NAMEEMBEDDING_DIMENSIONS— Embedding vector dimensions — must match the chosen model 384MAX_EMBED_CHARS— Maximum characters to embed per chunk 2000AGENT_PROFILE_ENGINE_ENABLEDVECT_DEBOUNCE_SECONDS— Debounce delay before vectorizing a changed file 0.5TASKS_DIR— /data # leave as-is for Docker; data persists in a named volumeLOG_LEVELHOSTPORT— HTTP server port 8000MAINTENANCE_INTERVAL_SECONDSMCP_DEBUG_TRANSPORTBASE_URLWORKER_OUTBOX_PATHWORKER_FLUSH_INTERVAL_SECONDSWORKER_FLUSH_CONCURRENCYMCP_SECRET_TOKEN[](https://m8ven.ai/mcp/desikai-lab-marrow-6yv107)