Home › Knowledge & Memory › mnemos
mnemos
Persistent memory with three swappable storage backends (Postgres/pgvector, Qdrant, Neo4j) behind one interface, benchmarked head-to-head for latency, not just claimed interchangeable. Retrieval weighting adapts per user via epsilon-greedy over real outcome feedback instead of a fixed formula. Reflection merges near-duplicate facts and decays/archives stale ones, audit-logged, never hard-deleted. Real before/after MCP eval: +60% answer accuracy with memory vs. without. `git clone` + `uv sync` (not yet on PyPI).
Topics: Knowledge & Memory
Connect
See the repository for installation instructions: https://github.com/vishalbanwari26/mnemos
Related servers
Context7
Up-to-date code docs for any prompt
cognee
Memory manager for AI apps and Agents using various graph and vector stores and allowing ingestion from 30+ data sources
hindsight
Hindsight: Agent Memory That Works Like Human Memory - Built for AI Agents to manage Long Term Memory
Skill_Seekers
Transform 17 source types (docs, GitHub repos, PDFs, videos, Jupyter, Confluence, Notion, Slack/Discord) into AI-ready skills and RAG…
basic-memory
Local-first knowledge management with bi-directional LLM sync via Markdown files.
NoteDiscovery
Self-hosted plain-markdown knowledge base with a built-in MCP server. Lets Claude Desktop, Cursor, and any MCP client search, read,…
Listed in punkpeye/awesome-mcp-servers (MIT)