qdrant-llamaindex-mcp-server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| QDRANT_URL | No | URL of the Qdrant server | |
| QDRANT_API_KEY | No | API key for the Qdrant server | |
| COLLECTION_NAME | No | Deprecated: Collection names are now specified dynamically by MCP clients at runtime | |
| EMBEDDING_MODEL | No | Default embedding model (used as fallback when auto-detection fails or model not in whitelist) | sentence-transformers/all-MiniLM-L6-v2 |
| QDRANT_READ_ONLY | No | Enable read-only mode (disables write tools for safety) | false |
| QDRANT_LOCAL_PATH | No | Path to the local Qdrant database (alternative to QDRANT_URL) | |
| EMBEDDING_PROVIDER | No | Embedding provider to use (currently only 'fastembed' is supported) | fastembed |
| TOOL_FIND_DESCRIPTION | No | Custom description for the find tool | |
| EMBEDDING_ALLOWED_MODELS | No | JSON array of allowed embedding models for dynamic loading | ["sentence-transformers/all-MiniLM-L6-v2", "BAAI/bge-small-en-v1.5", "snowflake/snowflake-arctic-embed-xs", "jinaai/jina-embeddings-v2-small-en"] |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| qdrant-findB | Search for documents stored by LlamaIndex in Qdrant. Use this tool when you need to:
|
| qdrant-get-pointB | Get a specific point by its ID from a Qdrant collection. |
| qdrant-get-collectionsA | Get a list of all collections in the Qdrant server. |
| qdrant-get-collection-detailsB | Get detailed information about a specific collection including status, configuration, and statistics. |
| qdrant-get-collection-countB | Get the number of points in a collection. |
| qdrant-peek-collectionB | Preview sample points from a collection. |
| qdrant-get-documentsC | Retrieve multiple documents by their IDs. |
| qdrant-search-by-vectorB | Search using a raw vector instead of text query. |
| qdrant-list-document-idsB | List document IDs with pagination support. |
| qdrant-scroll-pointsC | Paginated retrieval of points using scroll. |
| qdrant-storeB | Store information in Qdrant vector database. Use this tool when you need to:
|
| qdrant-delete-pointB | Delete a specific point by its ID from a Qdrant collection. |
| qdrant-update-point-payloadB | Update the payload (metadata) of a specific point by its ID. |
| qdrant-create-collectionB | Create a new collection with vector configuration. |
| qdrant-delete-collectionC | Delete a collection entirely. |
| qdrant-add-documentsC | Add multiple documents in batch. |
| qdrant-delete-documentsC | Delete multiple documents by their IDs. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 17 tools
Tools are mostly distinct with clear purposes across collections, documents, points, and search. Some overlap exists (e.g., delete-documents vs. delete-point, find vs. search-by-vector) but descriptions help differentiate.
Most tools follow a 'qdrant-verb-noun' pattern (e.g., add-documents, create-collection). However, 'qdrant-store' and 'qdrant-find' are verb-only, creating a slight inconsistency.
17 tools is on the higher side but still appropriate for a full-featured vector database server covering collection and document management, search, and maintenance. Each tool appears justified.
Core CRUD and search operations are covered, but missing tools for updating collection configuration or full document content updates (only payload update). Some gaps like truncating collections are absent.