OpenAI Assistants API Wrapper

Table of Contents

• OpenAI Assistants API Wrapper
  • Table of Contents
  • Introduction
  • Setup
  • API Endpoints
    • Assistants
    • Threads
    • Messages
    • Runs
    • Vector Stores
    • Files
    • Run Steps
    • Streaming
    • Function Calling (Tools)
  • Configuration
  • Error Handling
  • Security
  • Usage Examples
    • Assistants
      • Create Assistant
      • List Assistants
    • Threads
      • Create Thread
    • Messages
      • Create Message
    • Runs
      • Create Run
    • Streaming
      • Stream Run
    • Files
      • Upload File
    • Configuration
      • Update Rate Limit
      • Update File Size Limit
      • Get API Defaults
      • Set API Key

Introduction

This API wrapper provides a comprehensive interface to interact with OpenAI's Assistants API. It offers endpoints for managing assistants, threads, messages, runs, vector stores, and file operations, making it easier to integrate OpenAI's powerful AI capabilities into your applications.

Setup

1. Clone the repository:

   git clone https://github.com/RaheesAhmed/openai-assistants-api-v2.git
   cd openai-assistants-api-v2
   

2. Install dependencies:

   npm install
   

3. Set up environment variables in a .env file and add OPENAI_API_KEY other are not necessary:

   OPENAI_API_KEY=your_openai_api_key
   PORT=3000
   DEFAULT_ASSISTANT_ID=your_default_assistant_id
   DEFAULT_THREAD_ID=your_default_thread_id
   DEFAULT_MESSAGE_ID=your_default_message_id
   DEFAULT_RUN_ID=your_default_run_id
   DEFAULT_FILE_ID=your_default_file_id
   DEFAULT_VECTOR_STORE_ID=your_default_vector_store_id
   

4. Start the server:

   npm start
   

API Endpoints

Assistants

• Create Assistant: POST /create-assistant
• List Assistants: GET /list-assistants
• Retrieve Assistant: GET /retrieve-assistant/:assistant_id
• Update Assistant: PATCH /update-assistant/:assistant_id
• Delete Assistant: DELETE /delete-assistant/:assistant_id

Threads

• Create Thread: POST /create-thread
• Retrieve Thread: GET /retrieve-thread/:thread_id
• Update Thread: PATCH /update-thread/:thread_id
• Delete Thread: DELETE /delete-thread/:thread_id

Messages

• Create Message: POST /create-message/:thread_id
• List Messages: GET /list-messages/:thread_id
• Retrieve Message: GET /retrieve-message/:thread_id/:message_id
• Update Message: PATCH /update-message/:thread_id/:message_id

Runs

• Create Run: POST /create-run/:thread_id
• Retrieve Run: GET /retrieve-run/:thread_id/:run_id
• Cancel Run: POST /cancel-run/:thread_id/:run_id
• Submit Tool Outputs: POST /submit-tool-outputs/:thread_id/:run_id

Vector Stores

• Create Vector Store: POST /create-vector-store
• List Vector Stores: GET /list-vector-stores
• Retrieve Vector Store: GET /retrieve-vector-store/:vector_store_id
• Update Vector Store: PATCH /update-vector-store/:vector_store_id
• Delete Vector Store: DELETE /delete-vector-store/:vector_store_id

Files

• Upload File: POST /upload-file
• List Files: GET /list-files
• Retrieve File: GET /retrieve-file/:file_id
• Delete File: DELETE /delete-file/:file_id
• Retrieve File Content: GET /retrieve-file-content/:file_id

Run Steps

• List Run Steps: GET /list-run-steps/:thread_id/:run_id
• Retrieve Run Step: GET /retrieve-run-step/:thread_id/:run_id/:step_id

Streaming

• Stream Run: POST /stream-run/:thread_id/:run_id

Function Calling (Tools)

• Execute Tools: POST /execute-tools/:thread_id/:run_id

Configuration

• Update Rate Limit: POST /update-rate-limit
• Update File Size Limit: POST /update-file-size
• Get API Defaults: GET /api-defaults
• Set API Key: POST /set-api-key

Error Handling

The API uses a global error handling middleware to catch and process errors. Errors are returned in the following format:

{
  "error": {
    "message": "Error message"
  }
}

Security

• Uses helmet for setting various HTTP headers
• Implements rate limiting to prevent abuse
• Sanitizes file uploads and limits file size
• Uses CORS for cross-origin resource sharing

For production deployment, implement additional security measures such as authentication and input validation.

Usage Examples

Here are detailed examples of how to use the main endpoints:

Assistants

Create Assistant

const createAssistant = async () => {
  const response = await fetch('http://localhost:3000/create-assistant', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({
      name: "Math Tutor",
      instructions: "You are a helpful math tutor.",
      model: "gpt-4-1106-preview"
    })
  });
  const data = await response.json();
  console.log('Assistant created:', data);
};

List Assistants

const listAssistants = async () => {
  const response = await fetch('http://localhost:3000/list-assistants');
  const data = await response.json();
  console.log('Assistants:', data);
};

Threads

Create Thread

const createThread = async () => {
  const response = await fetch('http://localhost:3000/create-thread', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({
      messages: [{ role: "user", content: "Hello, I need help with math." }]
    })
  });
  const data = await response.json();
  console.log('Thread created:', data);
};

Messages

Create Message

const createMessage = async (threadId) => {
  const response = await fetch(`http://localhost:3000/create-message/${threadId}`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({
      role: "user",
      content: "What is the formula for the area of a circle?"
    })
  });
  const data = await response.json();
  console.log('Message created:', data);
};

Runs

Create Run

const createRun = async (threadId, assistantId) => {
  const response = await fetch(`http://localhost:3000/create-run/${threadId}`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ assistant_id: assistantId })
  });
  const data = await response.json();
  console.log('Run created:', data);
};

Streaming

Stream Run

const streamRun = async (threadId, runId) => {
  const response = await fetch(`http://localhost:3000/stream-run/${threadId}/${runId}`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ stream: true })
  });

  const reader = response.body.getReader();
  const decoder = new TextDecoder();

  while (true) {
    const { value, done } = await reader.read();
    if (done) break;
    console.log('Streamed data:', decoder.decode(value));
  }
};

Files

Upload File

const uploadFile = async (file) => {
  const formData = new FormData();
  formData.append('file', file);
  formData.append('purpose', 'assistants');

  const response = await fetch('http://localhost:3000/upload-file', {
    method: 'POST',
    body: formData
  });
  const data = await response.json();
  console.log('File uploaded:', data);
};

Configuration

Update Rate Limit

const updateRateLimit = async () => {
  const response = await fetch('http://localhost:3000/update-rate-limit', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ windowMs: 900000, max: 100 })
  });
  const data = await response.json();
  console.log('Rate limit updated:', data);
};

Update File Size Limit

const updateFileSizeLimit = async (fileSize) => {
  const response = await fetch('http://localhost:3000/update-file-size', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ fileSize })
  });
  const data = await response.json();
  console.log('File size limit updated:', data);
};

Get API Defaults

const getApiDefaults = async () => {
  const response = await fetch('http://localhost:3000/api-defaults');
  const data = await response.json();
  console.log('API defaults:', data);
};

Set API Key

const setApiKey = async (apiKey) => {
  const response = await fetch('http://localhost:3000/set-api-key', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ apiKey })
  });
  const data = await response.json();
  console.log('API key set:', data);
};

Remember to replace placeholder values (like asst_abc123, thread_abc123, etc.) with actual IDs when using these requests.