Complete REST API documentation for integrating Team-AI's multi-agent development platform into your applications.
All API requests must be made to the base URL. Use HTTPS for all requests.
Team-AI uses JWT bearer tokens. Log in via the web dashboard to obtain a token, then include it on every API request.
Log in with email/password or Google OAuth to receive a JWT. Pass it in the Authorization header on every request.
curl -X GET https://team-ai.andrastelabs.com/api/projects \
-H "Authorization: Bearer eyJhbGciOiJIUzI1NiIs..."Note: The Anthropic API key you configure in dashboard Settings authenticates calls to Claude, not to this API.
RESTful endpoints for managing projects, agents, and deployments.
/projects
Create a new Team-AI project
name and description are required. Stack is inferred from the description by the Project Manager agent.
{
"name": "Weather Dashboard",
"description": "React frontend + Node.js backend using OpenWeatherMap API",
"tech_preferences": { }
}Project starts in planning. Call POST /projects/{id}/kickoff after reviewing the contract preview to start the agents.
{
"success": true,
"project_id": "295f665a-a171-4cdd-851d-6dcf18bf3c6c",
"status": "planning",
"github_repo_url": "https://github.com/...",
"contract_preview": { ... }
}/projects
List all projects
No query parameters. Returns all projects owned by the authenticated user.
An array of project objects with live progress computed from tasks.
[
{
"project_id": "295f665a-...",
"name": "Weather Dashboard",
"status": "active",
"progress": 42
}
]/projects/:id
Retrieve a specific project
/projects/:id/kickoff
Start the 8-agent pipeline after confirming the contract preview.
POST /projects. Transitions status from planning to active.
/agents
List all active agents across the user's projects.
/agents/:agent_id/status
Get the live status of a specific agent instance.
/projects/:id/tasks
List all tasks for a project, including status and assigned agent.
API rate limits are applied per API key to ensure fair usage and system stability.
Projects per minute (per user)
5
Enforced at
POST /api/projects
Read endpoints (list, status, artifacts)
No limit
LLM cost is billed by Anthropic
to your own API key
If you exceed the 5-per-minute project creation limit, the API returns 429 Too Many Requests:
{
"detail": "Rate limit exceeded. Max 5 projects per minute. Try again in 60s."
}Standard HTTP status codes with detailed error responses for debugging.
| Status Code | Error Type | Description |
|---|---|---|
| 200 | OK | Request succeeded |
| 201 | Created | Resource successfully created |
| 400 | Bad Request | Invalid request parameters or malformed JSON |
| 401 | Unauthorized | Missing or invalid JWT bearer token |
| 403 | Forbidden | Insufficient permissions for the resource |
| 404 | Not Found | Requested resource does not exist |
| 429 | Too Many Requests | Rate limit exceeded, retry after delay |
| 500 | Internal Server Error | Server error, please retry or contact support |
| 503 | Service Unavailable | Service temporarily unavailable for maintenance |
FastAPI-style error response — a single detail field with the human-readable message:
{
"detail": "Invalid email or password"
}Follow these recommendations for optimal API integration and performance.
status field on project responses — agents run asynchronouslydetail message on error responses for human-readable contextLog in to obtain your JWT and start issuing API requests against the multi-agent platform.