CodingIT is an AI-powered software engineering platform that provides live code execution, file uploads, real-time chat capabilities, and workflow management. This comprehensive API documentation covers all 27 endpoints across 9 major functional areas.

# Import the collection
curl -o postman-collection.json https://raw.githubusercontent.com/Gerome-Elassaad/CodingIT/main/postman-collection.json

# Import the environment
curl -o postman-environment.json https://raw.githubusercontent.com/Gerome-Elassaad/CodingIT/main/postman-environment.json

Configure the following essential variables in Postman:

{
  "BASE_URL": "http://localhost:3000",
  "E2B_API_KEY": "your_e2b_api_key",
  "SUPABASE_URL": "your_supabase_url",
  "SUPABASE_ANON_KEY": "your_supabase_anon_key",
  "anthropic_api_key": "your_anthropic_key"
}

Most endpoints require Supabase authentication. Use the Bearer token format:

Authorization: Bearer {{supabase_access_token}}

GET /api/auth/github?code={auth_code}&state={csrf_state}

Description: Handles GitHub OAuth callback and automatically sets up webhooks for the first 3 repositories.

Parameters:

• code (required): Authorization code from GitHub
• state (required): CSRF protection state parameter
• error (optional): Error from GitHub OAuth

Response: Redirects to /settings/integrations with success or error parameters

Example:

GET /api/auth/github?code=abc123&state=xyz789
# Redirects to: /settings/integrations?success=github_connected

POST /api/auth/github/revoke
Content-Type: application/json

{
  "access_token": "github_access_token_here"
}

Description: Revokes GitHub access token and disconnects the integration.

Response:

{
  "success": true
}

POST /api/chat
Content-Type: application/json
Authorization: Bearer {{supabase_access_token}}

{
  "messages": [
    {
      "role": "user",
      "content": "Create a Python script that analyzes data from a CSV file"
    }
  ],
  "userID": "user_123",
  "teamID": "team_456",
  "template": "code-interpreter-v1",
  "model": "claude-3-sonnet-20240229",
  "config": {
    "model": "claude-3-sonnet-20240229",
    "apiKey": "your_anthropic_key",
    "temperature": 0.7,
    "maxTokens": 4000
  }
}

Description: Generate AI-powered code fragments using various LLM models with structured output.

Supported Models:

• claude-3-sonnet-20240229
• claude-3-haiku-20240307
• gpt-4
• gpt-3.5-turbo

Supported Templates:

• code-interpreter-v1 - Python data analysis
• nextjs-developer - Next.js applications
• vue-developer - Vue.js applications
• streamlit-developer - Streamlit dashboards
• gradio-developer - Gradio ML interfaces

Response: Streaming JSON with fragment schema

POST /api/chat/workflow

Description: Generates multi-step workflows with AI detection. The system automatically determines if a request should be a single fragment or broken into multiple workflow steps.

Key Features:

• Automatic workflow detection with confidence scoring
• Multi-step fragment breakdown
• Dependency management between steps
• Automatic workflow persistence

POST /api/code/execute
Content-Type: application/json

{
  "sessionID": "session_12345",
  "code": "import pandas as pd\ndf = pd.DataFrame({'x': [1,2,3], 'y': [4,5,6]})\nprint(df)"
}

Description: Execute code in isolated E2B sandboxes with comprehensive result capture.

Response:

{
  "results": [...],
  "stdout": ["DataFrame output..."],
  "stderr": [],
  "error": null
}

POST /api/sandbox
Content-Type: application/json

{
  "fragment": {
    "commentary": "Creating a data analysis script",
    "template": "code-interpreter-v1",
    "title": "Data Analysis",
    "description": "Analyze CSV data with pandas",
    "additional_dependencies": ["seaborn"],
    "has_additional_dependencies": true,
    "install_dependencies_command": "pip install seaborn",
    "port": null,
    "file_path": "analysis.py",
    "code": "import pandas as pd\nimport matplotlib.pyplot as plt\n# Analysis code here"
  },
  "userID": "user_123",
  "teamID": "team_456"
}

Description: Creates new E2B sandbox instances, installs dependencies, and executes fragments.

Execution Flow:

1. Create sandbox with specified template
2. Install additional dependencies (if any)
3. Write code files to sandbox
4. Execute code and return results

POST /api/terminal
Content-Type: application/json

{
  "command": "ls -la",
  "sbxId": "sandbox_id_here",
  "workingDirectory": "/home/user",
  "teamID": "team_456"
}

Description: Execute terminal commands in existing sandbox environments with 30-second timeout.

GET /api/files?sessionID=session_123&template=code-interpreter-v1

Description: Returns complete file tree structure for sandbox.

Response:

[
  {
    "name": "home",
    "isDirectory": true,
    "children": [
      {
        "name": "user",
        "isDirectory": true,
        "children": [
          {
            "name": "main.py",
            "isDirectory": false,
            "path": "/home/user/main.py"
          }
        ]
      }
    ]
  }
]

GET /api/files/content?sessionID=session_123&path=/home/user/main.py

POST /api/files/content
Content-Type: application/json

{
  "sessionID": "session_123",
  "path": "/home/user/new_file.py",
  "content": "print('Hello, World!')\n",
  "template": "code-interpreter-v1"
}

• GET/POST /api/files/sandbox - Connect to existing sandbox by ID
• GET /api/files/sandbox/list - List files in connected sandbox

POST /api/debug
Content-Type: application/json
Authorization: Bearer {{supabase_access_token}}

{
  "error": "NameError: name 'pandas' is not defined",
  "context": {
    "template": "code-interpreter-v1",
    "line_number": 5
  },
  "code": "import matplotlib.pyplot as plt\ndf = pandas.DataFrame({'x': [1,2,3]})"
}

Description: AI-powered error analysis with debugging suggestions and severity assessment.

Response:

{
  "analysis": "The error occurs because pandas is not imported...",
  "suggestions": [
    "Add 'import pandas as pd' at the top of your script",
    "Use 'pd.DataFrame' instead of 'pandas.DataFrame'"
  ],
  "severity": "medium",
  "category": "import_error"
}

GET /api/debug?session_id=debug_123
GET /api/debug  # List all active sessions

POST /api/deployments
Content-Type: application/json
Authorization: Bearer {{supabase_access_token}}

{
  "fragment": {
    "template": "nextjs-developer",
    "title": "Portfolio Site",
    "code": "export default function Home() { return <h1>Hello</h1> }",
    "port": 3000
  },
  "config": {
    "provider": "vercel",
    "environment": "production",
    "domain": "my-app.vercel.app"
  }
}

Supported Providers:

• vercel - Vercel deployments
• netlify - Netlify deployments
• aws - AWS deployments
• heroku - Heroku deployments

• GET /api/deployments - List deployment history
• GET /api/deployments/{id} - Get deployment status
• DELETE /api/deployments/{id} - Cancel deployment
• POST /api/deployments/{id}/rollback - Rollback deployment

GET /api/integrations/github/repos?page=1&per_page=30&sort=updated&type=owner
Authorization: Bearer {{supabase_access_token}}

Parameters:

• page - Page number (default: 1)
• per_page - Results per page (default: 30)
• sort - Sort order: created, updated, pushed, full_name
• type - Repository type: all, owner, member

GET /api/integrations/github/repos/octocat/Hello-World?path=README.md&ref=main

Features:

• Browse repository files and directories
• Support for different branches/refs
• Automatic token refresh handling
• Rate limit management

POST /api/workflows
Content-Type: application/json
Authorization: Bearer {{supabase_access_token}}

{
  "name": "Data Processing Pipeline",
  "description": "Complete data processing with validation and analysis",
  "fragments": [
    {
      "id": "node_1",
      "type": "fragment",
      "position": { "x": 100, "y": 100 },
      "data": {
        "template": "code-interpreter-v1",
        "title": "Data Ingestion",
        "code": "import pandas as pd\ndf = pd.read_csv('data.csv')"
      }
    }
  ],
  "connections": [
    {
      "id": "conn_1",
      "source": { "nodeId": "node_1", "portId": "output_1" },
      "target": { "nodeId": "node_2", "portId": "input_1" },
      "dataType": "object"
    }
  ],
  "variables": [
    {
      "name": "input_file",
      "type": "string",
      "default": "data.csv"
    }
  ],
  "triggers": [
    {
      "id": "trigger_1",
      "type": "manual",
      "config": {}
    }
  ]
}

POST /api/workflows/{workflow_id}/execute
Content-Type: application/json

{
  "inputData": {
    "input_file": "sample_data.csv",
    "output_format": "json"
  },
  "triggerType": "manual"
}

Workflow Features:

• Visual node-based workflow designer
• Fragment dependency management
• Variable passing between steps
• Multiple trigger types (manual, scheduled, webhook)
• Execution tracking and logging

• GET /api/workflows - List all workflows
• GET /api/workflows/{id} - Get workflow details
• PUT /api/workflows/{id} - Update workflow
• DELETE /api/workflows/{id} - Delete workflow
• GET /api/workflows/{id}/execute - List executions

POST /api/import-dataset
Content-Type: application/json

{
  "subset": "python"
}

Description: Import code datasets from HuggingFace for embeddings and semantic search.

Supported Subsets:

• python - Python code samples
• javascript - JavaScript code samples
• java - Java code samples
• cpp - C++ code samples

Process:

1. Downloads dataset from HuggingFace
2. Processes code into chunks
3. Generates embeddings using Xenova/all-MiniLM-L6-v2
4. Stores in Supabase vector database

POST /api/webhooks/github
Content-Type: application/json
X-GitHub-Event: push
X-GitHub-Delivery: 12345-abcdef
X-Hub-Signature-256: sha256=signature_here

{
  "ref": "refs/heads/main",
  "repository": {
    "full_name": "user/repo",
    "owner": { "login": "user" }
  },
  "commits": [...],
  "pusher": { "name": "user" }
}

Supported Events:

• push - Repository push events
• pull_request - Pull request events
• issues - Issue events
• ping - Webhook test events

{
  "error": "Error message",
  "code": "ERROR_CODE",
  "details": {
    "field": "Additional error details"
  }
}

Rate limits apply to endpoints without custom API keys:

• Default: 10 requests per day
• With API Key: No rate limits
• Headers: X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset

• E2B account and API key
• Supabase project setup
• GitHub OAuth app configured (optional)
• AI provider API keys (Anthropic, OpenAI, etc.)
• Postman installed

1. Generate a simple Python script:

   POST /api/chat
   # Body: Request Python data analysis script

2. Execute the generated code:

   POST /api/code/execute
   # Body: sessionID + generated code

3. List created files:

   GET /api/files?sessionID={{session_id}}

4. Create a workflow:

   POST /api/workflows
   # Body: Multi-step data processing workflow

5. Execute the workflow:

   POST /api/workflows/{{workflow_id}}/execute

Generate code → Execute in sandbox → Iterate based on results

Import data → Process → Analyze → Visualize → Export

Generate Next.js app → Test locally → Deploy to Vercel

Connect GitHub → Import repository → Analyze code → Generate improvements

Create workflow → Set triggers → Execute automatically → Monitor results

The platform supports multiple sandbox templates optimized for different use cases:

• code-interpreter-v1: Data science and analysis
• nextjs-developer: React/Next.js development
• vue-developer: Vue.js development
• streamlit-developer: ML dashboard creation
• gradio-developer: ML interface development

Multiple AI providers and models are supported:

Anthropic:

• claude-3-sonnet-20240229
• claude-3-haiku-20240307

OpenAI:

• gpt-4
• gpt-3.5-turbo

Google:

• gemini-pro
• gemini-pro-vision

Automatic webhook setup for GitHub repositories enables:

• Real-time code change notifications
• Automated testing and deployment
• Issue and PR tracking
• Integration with external CI/CD systems

• Store sensitive keys in environment variables
• Use separate keys for development/production
• Rotate keys regularly
• Never commit keys to version control

• Use Supabase Row Level Security (RLS)
• Implement proper session management
• Validate all user inputs
• Use HTTPS in production

• E2B provides isolated execution environments
• Automatic cleanup after timeout
• Resource limits prevent abuse
• Network isolation for sensitive operations

• Postman Collection
• OpenAPI Specification
• Environment Variables

• E2B Documentation
• Supabase Documentation
• GitHub API Documentation

• GitHub Repository
• Issues & Feature Requests

Last Updated: December 2024
API Version: 1.0.0
Total Endpoints: 27