step

pinecone

Execute Pinecone vector database operations.

Overview

Supports operations:
- upsert: Insert or update vectors
- query: Find similar vectors
- delete: Delete vectors by ID
- fetch: Retrieve vectors by ID
- update: Update vector metadata or values

Setup:
1. Create a Pinecone account at https://www.pinecone.io/
2. Create a project and index in the Pinecone console
3. Get your API key from the API Keys section
4. Note your environment (e.g., us-west1-gcp) from the console
5. Store credentials securely (e.g., environment variables: PINECONE_API_KEY, PINECONE_ENVIRONMENT)

API Key: Required. Get from Pinecone console > API Keys

Examples

Query similar vectors

Find most similar vectors to a query embedding

type: pinecone
api_key: ${env:PINECONE_API_KEY}
environment: us-west1-gcp
index_name: document-embeddings
operation: query
query_vector: [0.1, 0.2, 0.3, 0.4, 0.5]
top_k: 10
include_metadata: true
output_to: similar_docs

Upsert vectors with metadata

Insert or update vectors with associated metadata

type: pinecone
api_key: ${env:PINECONE_API_KEY}
environment: us-west1-gcp
index_name: product-embeddings
operation: upsert
vectors:
  - id: prod-123
    values: [0.1, 0.2, 0.3]
    metadata:
      title: Product A
      category: electronics
  - id: prod-124
    values: [0.4, 0.5, 0.6]
    metadata:
      title: Product B
      category: home
output_to: upsert_result

Query with metadata filter

Search vectors filtered by metadata conditions

type: pinecone
api_key: ${env:PINECONE_API_KEY}
environment: us-west1-gcp
index_name: articles
operation: query
query_vector: [0.15, 0.25, 0.35, 0.45]
top_k: 5
filter:
  category: technology
  published_year:
    $gte: 2023
include_metadata: true
output_to: filtered_results

Delete vectors by ID

Remove specific vectors from the index

type: pinecone
api_key: ${env:PINECONE_API_KEY}
environment: us-west1-gcp
index_name: embeddings
operation: delete
vector_ids:
  - vec-001
  - vec-002
  - vec-003
output_to: delete_result

Configuration

Parameter	Type	Required	Description
`type`	`string`	No	Step type identifier Default: `"pinecone"`
`api_key`	`string`	Yes	Pinecone API key for authentication
`environment`	`string`	Yes	Pinecone environment/region identifier (e.g., us-west1-gcp, us-east-1-aws)
`index_name`	`string`	Yes	Name of the Pinecone index to operate on
`operation`	`string`	Yes	Operation to perform: upsert (insert/update vectors), query (similarity search), delete (remove vectors), fetch (get by ID), or update (modify metadata) Options: `upsert`, `query`, `delete`, `fetch`, `update`
`vectors`	`string`	No	List of vector objects to upsert, each with id, values (array of floats), and optional metadata
`vector_ids`	`string`	No	List of vector IDs for delete or fetch operations
`query_vector`	`string`	No	Query vector values for similarity search (required for query operation)
`top_k`	`integer`	No	Number of most similar results to return (for query operation) Default: `10`
`filter`	`string`	No	Metadata filter to apply to query results (Pinecone filter syntax)
`include_metadata`	`boolean`	No	Whether to include metadata in query/fetch results Default: `true`
`include_values`	`boolean`	No	Whether to include vector values in query/fetch results Default: `false`
`namespace`	`string`	No	Namespace to use for operations (for data isolation within an index)
`timeout`	`integer`	No	HTTP request timeout in seconds Default: `30`
`output_to`	`string`	No	Key name where results will be stored in the event Default: `"pinecone"`

Base Configuration

These configuration options are available on all steps:

Parameter	Type	Default	Description
`name`		`null`	Optional name for this step (for documentation and debugging)
`description`		`null`	Optional description of what this step does
`retries`	`integer`	`0`	Number of retry attempts (0-10)
`backoff_seconds`	`number`	`0`	Backoff (seconds) applied between retry attempts
`retry_propagate`	`boolean`	`false`	If True, raise last exception after exhausting retries; otherwise swallow.

← Back to All Steps