pinecone

Pinecone is a vector database designed for building high-performance vector search applications. It enables efficient storage, management, and similarity search of high-dimensional vector embeddings, making it ideal for AI applications that require semantic search capabilities.

With Pinecone, you can:

• Store vector embeddings: Efficiently manage high-dimensional vectors at scale

• Perform similarity search: Find the most similar vectors to a query vector in milliseconds

• Build semantic search: Create search experiences based on meaning rather than keywords

• Implement recommendation systems: Generate personalized recommendations based on content similarity

• Deploy machine learning models: Operationalize ML models that rely on vector similarity

• Scale seamlessly: Handle billions of vectors with consistent performance

• Maintain real-time indexes: Update your vector database in real-time as new data arrives

The Agent Forge Pinecone integration allows your agents to programmatically use vector search, bringing sophisticated automation to your workflows that combines natural language processing with semantic search.

Your agents gain the ability to generate text embeddings, store them in Pinecone indexes, and perform similarity searches to find the most relevant information based on semantic meaning rather than simple keyword matching.

By connecting Agent Forge with Pinecone, you bridge the gap between AI workflows and vector search infrastructure, allowing you to create agents that understand context, retrieve relevant data from large datasets, and deliver more personalized, accurate responses without complex infrastructure management.

Usage Instructions

Store, search, and retrieve vector embeddings using Pinecone's specialized vector database. Generate embeddings from text and perform semantic similarity searches with customizable filtering options.

Where to get the Pinecone API key?

The API key is essential to use the tool. To get a Pinecone API key, you will need to use their web console.

Here is a step-by-step guide based on the official documentation:

1

Sign up or log in to the Pinecone console

2

Navigate to the "API Keys" section

Once you are logged in and have selected your project, find the "API Keys" tab in the left-hand navigation menu.

3

Create a new API key

• Click the "Create API key" button.

• Give your new key a descriptive name.

• (Optional) If you are on a paid plan, you can select custom permissions for the key. For the free Starter plan, the permissions are set to "All" by default.

4

Copy and save your key

• After you click "Create key," Pinecone will display the API key value.

• Important: This is the only time the full key will be shown. Copy it immediately and save it in a secure location, like a password manager or a .env file or in the Environment in Setings in Agent Forge, as you will not be able to retrieve it again.

• You will also need your environment name, which is typically found on the same "API Keys" page or in your project settings.

Tools

pinecone_generate_embeddings

Generate embeddings from text using Pinecone

Input

Parameter	Type	Required	Description
model	string	Yes	Model to use for generating embeddings
inputs	array	Yes	Array of text inputs to generate embeddings for
apiKey	string	Yes	Pinecone API key

Output

Parameter	Type	Description
matches	any	Search matches
upsertedCount	any	Upserted count
data	any	Response data
model	any	Model information
vector_type	any	Vector type
usage	any	Usage statistics

pinecone_upsert_text

Insert or update text records in a Pinecone index

Input

Parameter	Type	Required	Description
indexHost	string	Yes	Full Pinecone index host URL
namespace	string	Yes	Namespace to upsert records into
records	array	Yes	Record or array of records to upsert, each containing _id, text, and optional metadata
apiKey	string	Yes	Pinecone API key

Output

Parameter	Type	Description
matches	any	Search matches
upsertedCount	any	Upserted count
data	any	Response data
model	any	Model information
vector_type	any	Vector type
usage	any	Usage statistics

pinecone_search_text

Search for similar text in a Pinecone index

Input

Parameter	Type	Required	Description
indexHost	string	Yes	Full Pinecone index host URL
namespace	string	No	Namespace to search in
searchQuery	string	Yes	Text to search for
topK	string	No	Number of results to return
fields	array	No	Fields to return in the results
filter	object	No	Filter to apply to the search
rerank	object	No	Reranking parameters
apiKey	string	Yes	Pinecone API key

Output

Parameter	Type	Description
matches	any	Search matches
upsertedCount	any	Upserted count
data	any	Response data
model	any	Model information
vector_type	any	Vector type
usage	any	Usage statistics

pinecone_search_vector

Search for similar vectors in a Pinecone index

Input

Parameter	Type	Required	Description
indexHost	string	Yes	Full Pinecone index host URL
namespace	string	No	Namespace to search in
vector	array	Yes	Vector to search for
topK	number	No	Number of results to return
filter	object	No	Filter to apply to the search
includeValues	boolean	No	Include vector values in response
includeMetadata	boolean	No	Include metadata in response
apiKey	string	Yes	Pinecone API key

Output

Parameter	Type	Description
matches	any	Search matches
upsertedCount	any	Upserted count
data	any	Response data
model	any	Model information
vector_type	any	Vector type
usage	any	Usage statistics

pinecone_fetch

Fetch vectors by ID from a Pinecone index

Input

Parameter	Type	Required	Description
indexHost	string	Yes	Full Pinecone index host URL
namespace	string	No	Namespace to fetch vectors from
vectorIds	array	Yes	Array of vector IDs to fetch
apiKey	string	Yes	Pinecone API key

Output

Parameter	Type	Description
matches	any	Search matches
upsertedCount	any	Upserted count
data	any	Response data
model	any	Model information
vector_type	any	Vector type
usage	any	Usage statistics

Notes

• Category: tools

• Type: pinecone