n8n-nodes-qdrant-context-retriever
v1.0.3
Published
n8n community node for Qdrant — retrieves semantically relevant context from a Qdrant vector database using text embeddings, for RAG and semantic search workflows.
Maintainers
Readme
n8n-nodes-qdrant-context-retriever
An n8n community node that retrieves semantically relevant context from a Qdrant-backed vector search API, for RAG and semantic search workflows.
Installation
Follow the n8n community nodes installation guide, or install manually:
npm install n8n-nodes-qdrant-context-retrieverConfiguration
This node does not use a dedicated n8n credential type — connection details are set directly on the node:
| Field | Description |
|---|---|
| Vector API Host | Hostname of your Vector API service (defaults to vector-api, a container-internal service name) |
| Vector API Port | Port number of the Vector API service (container-internal, not a host-mapped port) |
| Bearer Token | Auth token sent as a Bearer token to the Vector API |
Note: because the Bearer Token is a plain node parameter rather than an n8n credential, it is stored with the workflow definition rather than in n8n's encrypted credential store, and will be visible to anyone with access to the workflow JSON. If this matters for your use case, consider wrapping it in an n8n HTTP Request credential or Generic Credential Type instead so it benefits from n8n's credential encryption and per-user access control.
Operation
The Qdrant Context Retriever node takes a User Intention / Query and returns matching context:
- Number of Results — max contexts to retrieve (1–50)
- Minimum Score Threshold — filters out low-relevance matches
- Output Format — JSON (structured, with metadata and scores), Markdown (formatted, with sources and scores), or Text (plain, no scores)
- Include Metadata — toggle inclusion of source/file-type metadata in results
License
MIT
