Omega

This is a local-first memory system built on SQLite that turns agent interactions into ranked, retrievable context. It exposes semantic search, FTS5 lookup, and relationship navigation through MCP, so Claude can store decisions, recall past corrections, and surface related knowledge clusters without cloud dependencies. The core handles capture, retrieval, and contradiction detection. The Pro tier adds multi-agent coordination with file claims and session rosters, LLM routing across providers, and knowledge base indexing for PDFs and markdown docs. Reach for this when you need session continuity across long-running projects or want agents to build on institutional knowledge instead of starting blank every time. Runs on Python 3.11+ with local embeddings and no upstream data sync.