Long-term Memory for AI. The Pinecone vector database makes it easy to build high-performance vector search applications. Developer-friendly, fully managed, and easily scalable without infrastructure hassles.
Long-term Memory for AI. The Pinecone vector database makes it easy to build high-performance vector search applications. Developer-friendly, fully managed, and easily scalable without infrastructure hassles. On Nagent, Pinecone is exposed as a fully-configurable artificial intelligence integration that any agent can call — 48 actions, and API_KEY, OAUTH2 authentication. No code is required to wire Pinecone into your workflow — connect it once via the External Integrations panel and reuse it across every agent you build.
Agent builders use Pinecone to automate the kinds of tasks artificial intelligence teams previously handled manually. Concrete examples — each one is a single agent step in Nagent — include:
Every action and trigger is paired with a structured input/output schema (visible in the sections below), so when you wire Pinecone into Helix — our agentic agent builder — the editor knows exactly what each step expects and produces. Configure once, deploy anywhere across your Nagent agents.
Every operation an agent can call against Pinecone, with input parameters and output schema. Drop these into any step of an agent built in Helix.
PINECONE_CANCEL_BULK_IMPORTTool to cancel a bulk import operation in Pinecone. Use when you need to stop an ongoing import operation that is not yet finished.
Input parameters
Unique identifier of the import operation to cancel
The host URL of the index where the bulk import is running. Get this from the list indexes API or Pinecone console (e.g., 'your-index-epf46k1.svc.aped-4627-b74a.pinecone.io')
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CHAT_ASSISTANTTool to chat with a Pinecone assistant and get structured responses with citations. Use when you need to query an assistant that has access to your knowledge base and want to get back answers with document references and citations.
Input parameters
The large language model to use for answer generation. Default is 'gpt-4o'.
Optionally filter which documents can be retrieved using metadata fields. Example: {"genre": {"$ne": "documentary"}}
If false, the assistant returns a single JSON response. If true, the assistant returns a stream of responses. Default is false.
List of messages in the conversation. Must include at least one message with role 'user'.
Controls randomness of model output: lower values make responses more deterministic, higher values increase creativity. Default is 0.0.
If true, the assistant will be instructed to return a JSON response. Cannot be used with streaming. Default is false.
The host URL for the assistant endpoint (e.g., 'prod-1-data.ke.pinecone.io'). This is the data plane host for your assistant.
The name of the assistant to chat with.
Controls the context snippets sent to the LLM.
If true, the assistant will be instructed to return highlights from the referenced documents that support its response. Default is false.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CHAT_COMPLETION_ASSISTANTTool to chat with a Pinecone assistant through an OpenAI-compatible interface. Use when you need to interact with a Pinecone assistant that has access to indexed documents and can answer questions based on retrieved context.
Input parameters
The large language model to use for answer generation.
Optionally filter which documents can be retrieved using metadata fields.
If false, returns a single JSON response. If true, returns a stream of responses.
The list of messages in the conversation, ordered from oldest to newest.
Controls the randomness of the model's output: lower values make responses more deterministic, while higher values increase creativity. Range: 0.0-2.0.
The host URL of the assistant. Get this from the list assistants operation.
The name of the assistant to chat with.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CONFIGURE_INDEXTool to configure an existing Pinecone index, including pod type, replicas, deletion protection, and tags. Use when you need to scale an index vertically or horizontally, enable/disable deletion protection, or update tags. The change is asynchronous; check index status for completion.
Input parameters
Index specification containing pod or serverless configuration.
Updates custom tags as key-value pairs. Set value to empty string to delete a tag. Keys max 80 chars, values max 120 chars.
The name of the index to configure.
Required date-based version header for the Pinecone API.
Sets deletion protection for the index. Use 'enabled' to prevent accidental deletion or 'disabled' to allow deletion.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CREATE_ASSISTANTTool to create a new Pinecone assistant for RAG (Retrieval-Augmented Generation) applications. Use when you need to initialize a new assistant that can have files uploaded and support chat interactions.
Input parameters
The name of the assistant. Resource name must be 1-63 characters long, start and end with an alphanumeric character, and consist only of lower case alphanumeric characters or '-'.
Optional metadata associated with the assistant. JSON object that can store custom organizational data, tags, and attributes.
Required date-based version header for the Pinecone API.
Description or directive for the assistant to apply to all responses. System-level instructions that guide the assistant's behavior.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CREATE_BACKUPTool to create a backup of a Pinecone index for disaster recovery and version control. Use when you need to preserve the current state of an index including vectors, metadata, and configuration.
Input parameters
User-defined identifier for the backup.
Name of the index to backup.
Contextual information about the backup purpose.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CREATE_INDEXTool to create a Pinecone index with specified configuration. Use when you need to initialize a new vector database index for storing and querying embeddings.
Input parameters
Index identifier (1-45 chars, lowercase alphanumeric or hyphens, must start/end with alphanumeric).
Deployment configuration. Must provide either serverless or pod specification.
Similarity measure. Must be 'dotproduct' for sparse vectors.
Vector dimensions (1-20,000). Required for dense vectors; omit for sparse vectors.
Vector type: 'dense' (default) or 'sparse'.
Set to 'enabled' to prevent accidental deletion.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CREATE_INDEX_FOR_MODELTool to create a Pinecone index with integrated embedding model for automatic vectorization. Use when you need to set up a new index that automatically converts text to vectors using a pre-configured embedding model.
Input parameters
Index identifier. Must be 1-45 characters, alphanumeric and hyphens only, must start and end with alphanumeric character.
Custom metadata labels. Keys max 80 characters, values max 120 characters.
Cloud provider for hosting the index.
Embedding model configuration for automatic vectorization.
Deployment region. Must be valid for the chosen cloud provider (e.g., 'us-east-1' for AWS).
Protection against accidental deletion. Defaults to 'disabled' if not specified.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CREATE_INDEX_FROM_BACKUPTool to create an index from a backup. Use when you need to restore or duplicate index data from a previously saved backup.
Input parameters
Index name, 1-45 characters, alphanumeric lowercase or hyphens only.
Custom key-value pairs for organization.
The ID of the backup to create an index from.
Enable or disable deletion safeguards.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_CREATE_NAMESPACETool to create a namespace within a serverless Pinecone index. Use when you need to organize vectors into isolated partitions.
Input parameters
The name of the namespace to create.
Schema configuration for metadata indexing behavior.
The host URL of the serverless index (e.g., 'index-name-xxxxx.svc.aped-xxxx-xxxx.pinecone.io').
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DELETE_ASSISTANTTool to permanently delete a Pinecone assistant. Use when you need to remove an assistant from your project.
Input parameters
The name of the assistant to delete.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DELETE_BACKUPTool to delete a backup. Use when you need to permanently remove a backup from your project.
Input parameters
The ID of the backup to delete.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DELETE_FILETool to delete an uploaded file from a Pinecone assistant. Use when you need to remove a file that was previously uploaded to an assistant.
Input parameters
The host URL of the assistant. This is typically in the format 'prod-1-data.ke.pinecone.io' or similar.
The name of the assistant to delete the file from.
The UUID of the file to be deleted.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DELETE_INDEXTool to permanently delete a Pinecone index. Use when you need to remove an index from your project. Note: Deletion protection and pending collections can prevent deletion.
Input parameters
The name of the index to delete. Example: 'test-index'
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DELETE_NAMESPACETool to permanently delete a namespace from a serverless index. Use when you need to remove an entire namespace and all its data. This operation is irreversible and only supported on serverless indexes.
Input parameters
The namespace identifier to be deleted. This operation is irreversible and will destroy all vectors in the namespace.
The host URL of the serverless index containing the namespace. Can be obtained from the list indexes operation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DELETE_VECTORSTool to delete vectors by ID from a Pinecone index. Use when you need to remove specific vectors from a namespace. Supports deletion by IDs, metadata filter, or entire namespace deletion.
Input parameters
List of vector IDs to delete. Mutually exclusive with deleteAll and filter. Length: 1-1,000 IDs.
Metadata filter to select vectors for deletion. Mutually exclusive with ids and deleteAll. Example: {"genre": {"$in": \["comedy", "documentary"\]}, "year": {"$eq": 2019}}
The namespace to delete vectors from. If not specified, uses the default namespace.
If true, deletes all vectors in the namespace. Mutually exclusive with ids and filter. Default: false.
The host URL of the index to delete vectors from. Get this from the list_indexes response (e.g., 'my-index-abc123.svc.region.pinecone.io').
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DESCRIBE_BACKUPTool to retrieve detailed information about a specific backup. Use when you need to check backup status, configuration, or metadata.
Input parameters
The unique identifier of the backup to retrieve
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DESCRIBE_BULK_IMPORTTool to describe a specific bulk import operation in Pinecone. Use when you need to retrieve detailed information about an import's status, progress, timing, and any errors.
Input parameters
Unique identifier for the import operation to describe.
The host URL of the Pinecone index where the bulk import is being performed. This is the data plane URL (e.g., 'my-index-abc123.svc.us-east1-gcp.pinecone.io').
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DESCRIBE_FILETool to get the status and metadata of a file uploaded to a Pinecone assistant. Use when you need to check file processing status, retrieve file metadata, or get a temporary signed URL for accessing the file contents.
Input parameters
Date-based version header for the Pinecone API.
Set to 'true' to include the signed URL of the file in the response. The signed URL provides temporary, read-only access to the underlying file.
The host URL of the assistant. This is typically in the format 'prod-1-data.ke.pinecone.io' or similar.
The name of the assistant to which the file was uploaded.
The UUID of the file to describe.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DESCRIBE_INDEXTool to retrieve full Pinecone index configuration including host and integrated embedding settings. Use when you need to check if an index has integrated inference (embed config) to avoid 'Integrated inference is not configured' errors by choosing the appropriate search method.
Input parameters
The name of the index to describe.
Date-based version header for the Pinecone API.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DESCRIBE_INDEX_STATSTool to get index statistics including vector count per namespace, dimensions, and fullness. Use when you need to understand the contents and status of an index.
Input parameters
Metadata filter expression to get stats for only vectors matching the filter. Only supported on pod-based indexes. Serverless and starter indexes do not support this parameter.
The host URL of the index to get stats for (e.g., 'index-name-xxx.svc.region.pinecone.io'). Get this from the list_indexes response.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DESCRIBE_NAMESPACETool to describe a namespace in a serverless index, including the total number of vectors. Use when you need to get metadata about a namespace such as record count and indexed fields.
Input parameters
The namespace to describe.
The host URL of the serverless index containing the namespace. Can be obtained from the list indexes operation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_DESCRIBE_RESTORE_JOBTool to get detailed information about a specific restore job in Pinecone. Use when you need to check the status, progress, or metadata of a restore operation.
Input parameters
The ID of the restore job to describe.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_EMBEDTool to generate vector embeddings for input text using Pinecone's hosted embedding models. Use when you need to convert text into vector representations for semantic search or similarity matching.
Input parameters
The embedding model to use (e.g., 'multilingual-e5-large').
List of input objects to generate embeddings for. Each item contains a text field.
Optional parameters for the embedding model.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_FETCH_VECTORSTool to fetch vectors by ID from a Pinecone index. Use when you need to retrieve vector data and metadata for specific vector IDs from a single namespace.
Input parameters
The vector IDs to fetch. Does not accept values containing spaces.
The namespace to fetch vectors from. If not provided, the default namespace is used.
The host URL of the Pinecone index to fetch vectors from. Get this from the list indexes operation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_GET_ASSISTANTTool to retrieve the status and details of a Pinecone assistant. Use when you need to check assistant configuration, deployment status, or metadata.
Input parameters
Required date-based version header for the Pinecone API.
The name of the assistant to get a status on. Resource name must be 1-63 characters long, start and end with an alphanumeric character, and consist only of lower case alphanumeric characters or '-'.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_GET_MODELTool to retrieve detailed information about a specific model hosted by Pinecone. Use when you need to understand model capabilities for embedding and reranking operations.
Input parameters
The name of the model to look up.
Required date-based version header for the Pinecone API.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_ASSISTANTSTool to list all assistants in a Pinecone project. Use when you need to retrieve all assistants with their configurations and status information.
Input parameters
The number of assistants to return per page. Must be between 1 and 100.
Token for retrieving the next page of results from a previous list operation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_BULK_IMPORTSTool to list all recent and ongoing bulk import operations in Pinecone. Use when you need to monitor or track the status of data import jobs. Supports pagination with a default limit of 100 imports per page.
Input parameters
Max number of operations to return per page. Range: 1-100, default: 100
The host URL of the index to list bulk imports for (e.g., 'index-name-abc123.svc.region.pinecone.io'). Get the host using the List Indexes action.
Pagination token to continue a previous listing operation
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_COLLECTIONSTool to list all collections in a Pinecone project (pod-based indexes only). Use when you need to view available collections.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_FILESTool to list all files in a Pinecone assistant with optional metadata filtering. Use when you need to retrieve files uploaded to a specific assistant or filter files by metadata properties.
Input parameters
The number of files to return per page. Must be between 1 and 100.
Optional JSON-encoded metadata filter for files. Use to filter files based on their metadata properties.
Date-based version header for the Pinecone API.
The host URL of the assistant (e.g., 'https://prod-1-data.ke.pinecone.io'). If not provided, will fetch the assistant details to get the host URL.
The name of the assistant to list files for.
Token for retrieving the next page of results from a previous list operation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_INDEX_BACKUPSTool to list all backups for a specific Pinecone index. Use when you need to view available backups for an index. Supports pagination via limit and paginationToken parameters.
Input parameters
The number of results to return per page. Must be between 1 and 100.
Name of the backed up index.
The token to use to retrieve the next page of results.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_INDEXESTool to list all indexes in a Pinecone project. Use when you need to retrieve all indexes with their configurations and status information.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_MODELSTool to list all available embedding and reranking models hosted by Pinecone. Use when you need to discover available models or filter by model type (embed/rerank) or vector type (dense/sparse).
Input parameters
Filter models by type. Use 'embed' for embedding models or 'rerank' for reranking models.
Filter embedding models by vector type. Only relevant when type='embed'. Use 'dense' for dense vectors or 'sparse' for sparse vectors.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_NAMESPACES_OPERATIONTool to list all namespaces in a serverless Pinecone index. Use when you need to discover available namespaces for data organization. Returns up to 100 namespaces by default with pagination support.
Input parameters
Maximum number of namespaces to return per page.
Prefix filter to return only namespaces that start with this prefix.
The host URL of the index to list namespaces from. You can get this from the list indexes API.
Required date-based version header for the Pinecone API.
Token for continuing a previous listing operation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_PROJECT_BACKUPSTool to list all backups for indexes in a Pinecone project. Use when you need to retrieve backup information across all project indexes. Supports pagination with limit and paginationToken parameters.
Input parameters
Number of results per page. Must be between 1 and 100. Default: 10.
Token for retrieving subsequent result pages. Obtained from the previous response.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_RESTORE_JOBSTool to list all restore jobs for a project with pagination support. Use when you need to view the status of restore operations or track restore progress.
Input parameters
Number of results to return per page. Range: 1-100, Default: 10.
Token for retrieving the next page of results.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_LIST_VECTORSTool to list vector IDs in a Pinecone serverless index. Use when you need to browse or retrieve vector identifiers from a namespace. Supports filtering by prefix and pagination for large result sets.
Input parameters
Maximum number of vector IDs to return per page. Default is 100.
Filter vector IDs by prefix. Does not accept values containing spaces.
The namespace to list vectors from. If not provided, the default namespace is used.
The host URL of the Pinecone index to list vectors from. Get this from ListIndexes action.
Pagination token to continue a previous listing operation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_QUERY_VECTORSTool to perform semantic search within a Pinecone index using a query vector. Retrieves IDs and similarity scores of the most similar items, ordered from most to least similar. Either vector or id parameter must be provided.
Input parameters
Vector ID to use as query. Maximum length is 512 characters. Required if vector is not provided.
Number of results to return per query. Must be between 1 and 10,000.
Metadata filtering criteria to constrain results. Example: {"genre": {"$in": \["comedy", "documentary"\]}, "year": {"$eq": 2019}}
Query vector values matching index dimensionality. Length must be between 1 and 20,000. Required if id is not provided.
The host URL of the Pinecone index to query. Get this from the list indexes operation.
The namespace to search. If omitted, searches the default namespace.
Sparse vector representation.
Whether to include vector data in the response. Default is false.
Whether to include metadata in the response. Default is false.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_RERANKTool to rerank documents by semantic relevance to a query. Use when you need to order retrieved documents by their semantic relevance to a user's search query using Pinecone's hosted reranking models.
Input parameters
The reranking model to apply (e.g., 'bge-reranker-v2-m3').
The search query against which documents are ranked.
Number of ranked results to return. Defaults to the total number of input documents if not specified.
Array of document objects requiring reranking.
Additional model-specific options. Refer to model documentation for supported parameters.
Fields to consider for reranking. Defaults to \['text'\]. Model-specific limits may apply.
Include documents in response. Defaults to true.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_RETRIEVE_CONTEXT_ASSISTANTTool to retrieve context snippets from a Pinecone assistant for RAG or agentic workflows. Use when you need to fetch relevant document snippets based on a query or conversation messages to provide context for language model responses.
Input parameters
The query to generate context for. Exactly one of query or messages must be provided.
Maximum number of context snippets to return. Default is 16. Maximum is 64.
Metadata filter to constrain which documents can be retrieved. Example: {"genre": {"$ne": "documentary"}}
The list of messages to use for generating context. Exactly one of query or messages must be provided.
Whether to retrieve image-related context snippets. If false, only text snippets are returned. Default is true.
Maximum context snippet size in tokens. Default is 2048. Minimum is 512. Maximum is 8192.
The host URL for the assistant (e.g., 'prod-1-data.ke.pinecone.io'). Get this from the assistant configuration.
The name of the assistant to retrieve context from.
If image-related context snippets are returned, this determines whether to include base64 image data. If false, only image captions are returned. Only available when multimodal=true. Default is true.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_SEARCH_RECORDS_NAMESPACETool to search records within a Pinecone namespace using text, vector, or ID query. Use when you need to find similar records based on embeddings or record IDs. Results can optionally be reranked for relevance.
Input parameters
Query configuration including top_k and optional filters, vector, or ID
Fields to include in results
Reranking configuration for search results.
The namespace to search within
The host URL for the Pinecone index (e.g., 'my-index-abc123.svc.pinecone.io'). Get this from listing indexes.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_START_BULK_IMPORTTool to start an asynchronous bulk import of vectors from object storage (S3, GCS, or Azure Blob Storage) into a Pinecone index. Use when you need to import large volumes of vectors from external storage. Returns an import ID to track the operation status.
Input parameters
The URI of the bucket/container and import directory containing data. Format: s3://BUCKET_NAME/IMPORT_DIR, gs://BUCKET_NAME/IMPORT_DIR, or Azure Blob Storage path.
Error handling mode for bulk import operation.
The host URL of the Pinecone index to import into. Get this from the index details (e.g., 'my-index-abc123.svc.aped-4627-b74a.pinecone.io'). Use List Indexes to find available index hosts.
The identifier of the storage integration used to access the data.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_UPDATE_ASSISTANTTool to update an assistant's instructions and configuration in Pinecone. Use when you need to modify an existing assistant's behavior by changing its instructions or metadata. Updates are applied immediately.
Input parameters
Optional metadata associated with the assistant. Metadata is a JSON object that can store custom organizational data, tags, and attributes. Maximum size is 16KB.
Required date-based version header for the Pinecone API.
Description or directive for the assistant to apply to all responses. Maximum 16 KB.
The name of the assistant to update.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_UPDATE_VECTORTool to update a vector in Pinecone by ID. Use to overwrite vector values and/or metadata. Supports bulk updates via metadata filters.
Input parameters
Vector's unique identifier. Length: 1-512 characters.
If true, return the number of records matching the filter without executing the update. Default: false.
Metadata filter expression for bulk updates. Example: {"genre": {"$in": \["comedy", "documentary"\]}, "year": {"$eq": 2019}}
Vector data to update. Length: 1-20,000 elements. If provided, overwrites previous values.
The host URL of the index to update vectors in. Get this from the list_indexes response (e.g., 'my-index-abc123.svc.region.pinecone.io').
The namespace containing the vector to update. If not specified, uses the default namespace.
Metadata to set for the vector. Key-value pairs that will be added or overwrite previous metadata.
Sparse vector representation.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_UPLOAD_FILETool to upload a file to a Pinecone assistant for processing. Use when you need to add documents, PDFs, or other files to an assistant for knowledge retrieval or multimodal processing.
Input parameters
The file to upload.
Optional JSON-encoded metadata for the file. Use for filtering or categorization.
Optional flag to opt in to multimodal file processing (PDFs only). Can be 'true' or 'false'. Default is 'false'.
Model representing a file with direct content.
The host URL of the assistant (e.g., 'https://prod-1-data.ke.pinecone.io'). If not provided, will use the base_url from metadata.
The name of the assistant to upload the file to.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_UPSERT_RECORDS_NAMESPACETool to upsert text records into a Pinecone namespace. Use when you need to add or update records with automatic text-to-vector conversion.
Input parameters
List of records to upsert. Each record must have an '_id' field and can contain arbitrary additional fields like 'chunk_text' for text to be converted to vectors.
The namespace to upsert records into.
The host URL of the index to upsert records into. Get this from ListIndexes action (e.g., 'test-embed-idx-99999-epf46k1.svc.aped-4627-b74a.pinecone.io').
Pinecone API version header. Version 2025-04 or later is required to use the '__default__' namespace.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PINECONE_UPSERT_VECTORSTool to upsert vectors into a Pinecone namespace, overwriting existing vector IDs. Use when you need to add or update vectors with their dense and/or sparse values and metadata.
Input parameters
Array of vectors to upsert. Recommended batch limit is up to 1000 vectors.
The namespace where you upsert vectors. If not specified, uses the default namespace.
The host URL of the index to upsert vectors into. Get this from ListIndexes action (e.g., 'test-embed-idx-01-epf46k1.svc.aped-4627-b74a.pinecone.io').
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
No publicly available marketplace agent is found using this tool yet. There are 61 agents privately built on Nagent that already use Pinecone.
Build on Nagent
Connect Pinecone to any Nagent agent in minutes — no API key management, no boilerplate. Just configure and deploy.
The five questions agent builders ask before adopting a new integration.
Open the External Integrations panel inside Nagent (app.nagent.ai/externalIntegration), find Pinecone, and click "Connect Now." You'll authenticate with API_KEY, OAUTH2 — Nagent handles credential storage and refresh automatically. Once connected, Pinecone is available to any agent in your workspace.
No. Nagent provides no-code integration for every tool. Once Pinecone is connected, you configure its 48 actions directly in the agent builder UI — no API calls, no boilerplate, no schema management.
Helix — Nagent's agentic agent builder — lets you drop Pinecone steps into any workflow visually. Pick an action (e.g., one of those listed above), fill in the inputs (Helix knows the required vs. optional schema for each parameter), and connect it to upstream/downstream steps. Triggers run as the entry point of an agent, so when a Pinecone event fires, the agent kicks off automatically.
Every Pinecone action and trigger ships with a fully-typed schema — input parameters with name, type, required flag, and description, plus the output payload shape. The schemas are documented in the sections above. Helix uses these schemas to validate your configuration at build time and to type-check the data flowing between steps.
Yes. While Pinecone ships with 48 pre-built artificial intelligence actions, you can layer custom logic around them inside Helix — pre/post-processing steps, conditional branches, retries, or stitching Pinecone together with other connected tools. For deeper customization, talk to our team about Nagent's Agentic AI Lab — forward-deployed engineers who build Pinecone-based workflows tailored to your business.