MCP Service Integration — Pinecone Vector Database
Provides efficient vector retrieval capabilities for Pinecone via the MCP protocol, enabling functions such as setting up indexes, inserting data, or text searching.
Environment Variables
- PINECONE_API_KEY needs to be set to your Pinecone API key
Usage Instructions
Pinecone MCP Server Reference Documentation
🗺️ Feature List
Tool Identifier | Function Description | Core Parameters |
---|---|---|
search-docs | Search Pinecone official documentation to answer related questions | query: query text |
list-indexes | List all Pinecone indexes | No parameters required |
describe-index | Describe the configuration of an index | name (index name) |
describe-index-stats | Provides data statistics for an index, including record count and available namespaces | name (index name) |
create-index-for-model | Create a new index using an integrated inference model to embed text into vectors | name (index name), embed configuration (refer to EmbedOption) |
upsert-records | Insert or update records in an index using an integrated inference model for text embedding | name (index name), namespace (namespace, user-specified), records (array of records), example: [{ "_id": "rec1", "chunk_text": "The Eiffel Tower was completed in 1889 and stands in Paris, France.", "category": "history" }] |
search-records | Search records in an index using an integrated inference model for text embedding | name (index name), namespace (namespace), query (query configuration) refer to QueryOption |
EmbedOption
Parameter ID | Parameter Description | |
---|---|---|
model | Embedding model | multilingual-e5-large, llama-text-embed-v2, pinecone-sparse-english-v0 |
fieldMap | Field mapping for embedding | The field name in the data record that contains the text content to be embedded. Records in the index must contain this field. Example: { text: 'chunk_text' }, where 'chunk_text' is the field name |
QueryOption
Parameter ID | Parameter Description | |
---|---|---|
topK | Top K most relevant records | Number |
inputs | Text string for retrieval | example { text: 'retrieval content' } |
Repository URL
https://github.com/pinecone-io/pinecone-mcp