Updated Blackbox AI Chat API Proxy (Reverse-Engineered)

This Express.js server acts as a proxy to interact with the Blackbox AI chat service via reverse-engineered endpoints. It provides advanced customization options including custom system prompts, image generation mode, web search integration, chat history with context, and support for multiple AI models.

⚠️ Warning: This is an unofficial, reverse-engineered implementation. Use at your own discretion and ensure compliance with Blackbox AI's terms of service and applicable laws.

Features

• Custom System Prompts: Tailor AI behavior with specific instructions
• Image Generation Mode: Toggle image generation capabilities
• Web Search Mode: Enable web search for enriched responses
• Chat History: Maintain conversation context with message history
• Multiple AI Models: Access various scraped model endpoints
• Custom Chat ID: Assign unique conversation identifiers
• Response Time Tracking: Measure API latency
• Robust Error Handling: Comprehensive error management

Prerequisites

• Node.js (v14 or higher recommended)
• npm (comes with Node.js)
• Knowledge of Blackbox AI's scraped API endpoints

Installation

1. Clone the repository:

git clone https://github.com/notsopreety/blackbox-api-v2
cd blackbox-api-v2

2. Install dependencies:

npm install

3. Configure environment variables (optional): Create a .env file:

PORT=3000
# Add custom headers or cookies if required

Usage

1. Start the server:

npm start

2. API available at http://localhost:3000 (or your configured PORT).

Docker Deployment

Using Docker

1. Build the Docker image:

docker build -t blackbox-api-v2 .

2. Run the container:

docker run -p 3000:3000 -d --name blackbox-api blackbox-api-v2

Using Docker Compose

1. Run the application with Docker Compose:

docker-compose up -d

2. To stop the application:

docker-compose down

Cloud Deployment

The Docker configuration is compatible with most cloud platforms that support Docker containers:

• AWS ECS/Fargate: Upload the image to ECR and deploy as a service
• Google Cloud Run: Upload the image to GCR and deploy as a serverless container
• Azure Container Apps: Upload the image to ACR and deploy as a container app
• DigitalOcean App Platform: Connect your repository with the Dockerfile
• Heroku: Use the Heroku Container Registry to deploy

Environment Variables

When deploying to cloud environments, you may need to configure these environment variables:

• PORT: The port on which the API will run (default: 3000)

API Endpoint

POST /api/chat

Request Body Parameters

Parameter	Type	Required	Default	Description
messages	Array	Yes	N/A	Array of message objects with role (user/assistant) and content. Must contain at least one message.
modelId	String	Yes	N/A	ID of the AI model endpoint to use. Must match a supported model.
userSystemPrompt	String	No	null	Custom prompt to define AI behavior. If null, uses default AI behavior.
webSearchModePrompt	Boolean	No	false	Enable/disable web search integration.
imageGenerationMode	Boolean	No	false	Enable/disable image generation.
chatId	String	No	"SamirXYZ"	Custom identifier for the chat session.

Required Data

The API requires specific data to function correctly:

• messages:
  • Must be an array
  • Each object must have:
    • role: String ("user" or "assistant")
    • content: String (message text)
  • Minimum one message required
  • Example: [{"role": "user", "content": "Hello"}]
• modelId:
  • Must be a string
  • Must match one of the supported model IDs (see table below)
  • Case-sensitive
  • Example: "deepseek-chat"

Failure to provide these required fields will result in an error response.

Supported Models (Scraped)

Model ID	Display Name
meta-llama/Llama-3.3-70B-Instruct-Turbo	Meta-Llama-3.3-70B-Instruct-Turbo
deepseek-chat	DeepSeek-V3
deepseek-reasoner	DeepSeek-R1
deepseek-ai/deepseek-llm-67b-chat	DeepSeek-LLM-Chat-(67B)

Example Requests

Minimal Required Request

{
  "messages": [
    {
      "role": "user",
      "content": "Tell me about space travel"
    }
  ],
  "modelId": "deepseek-chat"
}

• Meets minimum requirements
• Uses defaults: userSystemPrompt: null, webSearchModePrompt: false, imageGenerationMode: false, chatId: "SamirXYZ"

Chat with History

{
  "messages": [
    {
      "content": "hey im samir",
      "role": "user"
    },
    {
      "content": "Hey Samir! I'm Nova. How can I assist you today?",
      "role": "assistant"
    },
    {
      "content": "what did you say to my first response",
      "role": "user"
    }
  ],
  "modelId": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
  "chatId": "pVwOVmZ",
  "userSystemPrompt": "your name is nova",
  "webSearchModePrompt": true,
  "imageGenerationMode": false
}

• Shows required messages with history
• Includes optional parameters

Full Featured Request

{
  "messages": [
    {
      "role": "user",
      "content": "Design a futuristic city"
    }
  ],
  "modelId": "deepseek-reasoner",
  "chatId": "CityDesign001",
  "userSystemPrompt": "You are an urban planning AI with creative vision",
  "webSearchModePrompt": true,
  "imageGenerationMode": true
}

• Uses all features with required data

Response Format

Success Response (Chat History Example)

{
  "status": "success",
  "response": "Hey Samir! To your first message, I said: 'Hey Samir! I'm Nova. How can I assist you today?'",
  "model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
  "id": "pVwOVmZ",
  "responseTime": 1.3,
  "history": [
    {
      "content": "hey im samir",
      "role": "user"
    },
    {
      "content": "Hey Samir! I'm Nova. How can I assist you today?",
      "role": "assistant"
    },
    {
      "content": "what did you say to my first response",
      "role": "user"
    },
    {
      "content": "Hey Samir! To your first message, I said: 'Hey Samir! I'm Nova. How can I assist you today?'",
      "role": "assistant"
    }
  ]
}

Error Response (Missing Required Data)

{
  "status": "error",
  "response": null,
  "model": "Unknown Model",
  "id": "SamirXYZ",
  "responseTime": 0,
  "history": [],
  "error": {
    "message": "Messages array is required",
    "details": null
  }
}

Advanced Features

Custom System Prompts

• Default: null (standard AI behavior)
• Usage: String, e.g., "your name is nova"
• Purpose: Defines AI personality/response style

Image Generation Mode

• Default: false
• Usage: imageGenerationMode: true
• Purpose: Enables image-related responses (scraped API dependent)

Web Search Mode

• Default: false
• Usage: webSearchModePrompt: true
• Purpose: Adds web data to responses

Chat History

• Required: At least one message in messages
• Usage: Include prior messages for context
• Purpose: Enables multi-turn conversations

Custom Chat ID

• Default: "SamirXYZ"
• Usage: Any string, e.g., "pVwOVmZ"
• Purpose: Tracks conversation threads

Configuration

Modify these in the code:

• DEFAULT_SYSTEM_PROMPT: null
• MAX_TOKENS: 1024 (response length)
• VALIDATION_TOKEN: Scraped API token
• API_ENDPOINT: Reverse-engineered endpoint
• MODEL_CONFIG: Model mappings

Security Considerations

• 10MB body size limit
• Uses scraped headers/cookies
• Add authentication/rate limiting for production
• Use HTTPS in production

Development

Run locally:

npm start

Contributing

1. Fork the repository
2. Create feature branch (git checkout -b feature/improvement)
3. Commit changes (git commit -m 'Add improvement')
4. Push branch (git push origin feature/improvement)
5. Open Pull Request

Legal Disclaimer

This reverse-engineered Blackbox AI API implementation is for educational/research purposes only. Not affiliated with Blackbox AI. Usage may:

• Violate terms of service
• Change unexpectedly
• Require authorization

Use responsibly and comply with all laws/terms. Maintainers are not liable for consequences.

License

MIT License

Author

Samir Thakuri

• GitHub: notsopreety
• Email: itssamir444@gmail.com

Feel free to reach out for any queries or collaboration opportunities!

Support

If you encounter any issues or have questions, please:

1. Check the issues page
2. Create a new issue if needed
3. Provide detailed information about the problem