Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.asteragents.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

The Pinecone integration enables your agents to perform semantic search on vector embeddings stored in your Pinecone database. This allows agents to find relevant documents, build RAG (Retrieval-Augmented Generation) applications, and perform similarity searches based on meaning rather than keywords.

​
Features

  • Semantic Search: Find documents based on meaning and context
  • Vector Similarity: High-performance similarity matching using embeddings
  • Namespace Support: Search within specific data collections
  • Configurable Results: Control the number of results returned
  • Metadata Access: Retrieve document metadata along with similarity scores
  • RAG Applications: Perfect for building retrieval-augmented generation systems

​
Prerequisites

  • Active Pinecone account with a configured index
  • Pinecone API key with read permissions
  • Pinecone host URL for your index
  • Vector embeddings already stored in your index
  • Understanding of your namespace structure

​
Setup Guide

1

Enable Pinecone Integration

Navigate to Control Hub > Integrations and locate the Pinecone card. Click “Connect” to begin setup.
2

Configure Connection

Provide your Pinecone connection details:
  • API Key: Your Pinecone API key from the console
  • Host: Your Pinecone index host URL (e.g., your-index-abc123.svc.us-east1-gcp.pinecone.io)
3

Test Connection

Click “Test Connection” to verify your configuration. The system will attempt to connect and validate your index access.

​
Finding Your Pinecone Details

​
API Key

  1. Log into your Pinecone console
  2. Navigate to “API Keys” in the sidebar
  3. Copy your API key value

​
Host URL

  1. Go to your Pinecone console
  2. Select your index
  3. Copy the host URL from the index details (without https://)

​
Security Considerations

  • Store API keys securely and rotate them regularly
  • Use read-only API keys when possible
  • Implement proper access controls for sensitive data
  • Monitor query usage and costs
  • Consider using namespaces to isolate different data types
  • Regularly audit which agents have access to which namespaces

​
Troubleshooting

  • Verify your API key is correct and active
  • Check that the host URL is properly formatted
  • Ensure your Pinecone index is running
  • Confirm network connectivity to Pinecone
  • Verify your API key hasn’t expired
  • Check that the API key has proper permissions
  • Ensure you’re using the correct Pinecone project
  • Confirm the index exists in your account
  • Verify the namespace exists and contains data
  • Check that your index has vector embeddings
  • Ensure the query format is correct
  • Confirm the index dimensions match your embeddings