MCP Server
Retrieval & Memory
Qdrant MCP
1Semantic 2memory 3backed 4by 5Qdrant 6vectors
the six Ws · specification
W1
Who
Maintained officially by Qdrant.
W2
What
It lets an agent store and semantically retrieve information as vectors in a Qdrant collection.
W3
Where
It runs locally as a Python server connecting to a local or cloud Qdrant instance.
W4
When
An agent reaches for it when it needs a persistent semantic memory layer.
W5
Why
It provides drop-in vector storage and recall for agent memory.
W6
With
Python, a Qdrant URL (and API key for cloud), and an embedding model.
W7
Watch
Official vendor maintainer; read/write access is limited to the collection you configure, and it embeds text before storage. Outbound calls go to your Qdrant instance and chosen embedding provider.
alternatives
works with