Pinecone Assistant - Flows App

A Flows app that integrates with Pinecone's Assistant API, enabling you to create AI assistants with context retrieval, file management, and chat capabilities.

Quick Start

1. Install dependencies - npm install
2. Get Pinecone API Key - Visit https://app.pinecone.io/ and create an API key
3. Configure - Add your API key when installing the app in Flows
4. Deploy - Use the blocks to create assistants and manage your knowledge base

App Structure

├── main.ts                    # App definition and configuration
├── package.json               # Dependencies and scripts
├── tsconfig.json             # TypeScript configuration
├── blocks/                   # Block implementations
│   ├── assistant.ts          # Assistant resource management
│   ├── dataFile.ts          # Data file resource management
│   ├── simpleChat.ts        # Simple chat with assistant
│   ├── rawChat.ts           # Raw chat interface
│   ├── retrieveSnippets.ts  # Context retrieval from knowledge base
│   ├── uploadFile.ts        # Upload files to assistant
│   ├── deleteFile.ts        # Delete files from assistant
│   ├── listFiles.ts         # List assistant files
│   ├── updateAssistant.ts   # Update assistant configuration
│   └── shared/              # Shared utilities
│       └── chatHelpers.ts   # Chat helper functions
└── README.md                # This file

Features

Available Blocks

Resources

• Assistant - Create and manage Pinecone assistants
• Data File - Manage data files for your assistants

Actions

• Simple Chat - Chat with your assistant using simple text input/output
• Raw Chat - Advanced chat interface with full message control
• Context Retrieval - Retrieve relevant snippets from your knowledge base
• Upload File - Add files to your assistant's knowledge base
• Delete File - Remove files from your assistant
• List Files - View all files associated with an assistant
• Update Assistant - Modify assistant configuration and instructions

Configuration

The app requires a single configuration parameter:

• Pinecone API Key (required, sensitive) - Your API key from https://app.pinecone.io/

The app automatically validates your API key on installation by attempting to list your assistants.

Usage Examples

Basic Workflow

1. Create an Assistant

   • Use the Assistant resource block to create a new Pinecone assistant
   • Configure its instructions and model preferences

2. Upload Knowledge Base

   • Use Upload File to add documents to your assistant's knowledge base
   • Files are automatically processed and indexed by Pinecone

3. Chat with Context

   • Use Simple Chat for basic conversations
   • Use Context Retrieval to get relevant snippets before chatting
   • Use Raw Chat for advanced message handling

4. Manage Files

   • Use List Files to see all uploaded documents
   • Use Delete File to remove outdated content
   • Use Update Assistant to modify instructions or settings

Integration Patterns

The blocks are designed to work together in flows:

• Context → Chat: Retrieve relevant context before sending chat messages
• Upload → List: Upload files and verify they were added successfully
• Assistant → Update: Create an assistant then modify its configuration
• Chat → Context: Use chat responses to guide context retrieval

Development

Prerequisites

• Node.js 20+
• npm

Setup

npm install

Available Scripts

npm run typecheck    # Type checking
npm run format       # Code formatting

Testing Your App

1. Run type checking: npm run typecheck
2. Format code: npm run format

Note: TypeScript may show SDK import errors during development. This is expected - the SDK will be available when the app runs in the Flows environment.

Architecture

Pinecone Integration

This app integrates with Pinecone's Assistant API, which provides:

• Managed AI Assistants - Pre-configured AI models with custom instructions
• Knowledge Base - Automatic file processing and vector indexing
• Context Retrieval - Semantic search across your documents
• Chat Interface - Conversational AI with RAG (Retrieval Augmented Generation)

Block Design

Each block is designed for a specific use case:

• Resource blocks (Assistant, Data File) manage Pinecone entities
• Action blocks provide specific functionality like chat or file operations
• Shared utilities in blocks/shared/ provide common functionality

Error Handling

The app follows Flows best practices:

• Errors bubble up naturally without wrapping
• Descriptive error messages for debugging
• Automatic API key validation on app installation

Troubleshooting

Common Issues

"Invalid API Key" error

• Verify your API key is correct at https://app.pinecone.io/
• Check that the key has appropriate permissions
• Ensure the key is properly pasted without extra spaces

Assistant not found

• Make sure the assistant exists in your Pinecone account
• Verify you're using the correct assistant ID
• Check that the assistant is properly configured

File upload failures

• Ensure file size is within Pinecone's limits
• Check file format is supported
• Verify sufficient quota in your Pinecone account

Type checking errors

• Run npm run typecheck to identify specific issues
• Ensure all dependencies are installed
• Check that SDK imports are correct

Dependencies

• @pinecone-database/pinecone - Official Pinecone JavaScript SDK
• @slflows/sdk - Flows platform SDK
• TypeScript - Type safety and development tooling

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Run npm run typecheck and npm run format
5. Submit a pull request

License

This project is licensed under the MIT License - see the LICENSE file for details.