REST API

Base URL

All API requests go to:

https://api.forjegames.com

The API is versioned with a /v1 prefix. We ship non-breaking additions freely — breaking changes will be released under /v2 with a deprecation window of at least 6 months.

Authentication

Create an API key at Settings → API Keys. Pass it as a Bearer token in the Authorization header:

1curl https://api.forjegames.com/v1/me \
2  -H "Authorization: Bearer fjg_live_xxxxxxxxxxxxxxxxxxxx"

Rate limits

Default limits are 60 requests/minute per API key on the Creator plan, and 600 requests/minute on Studio. Every response includes:

• X-RateLimit-Limit — your current per-minute quota
• X-RateLimit-Remaining — requests left in the current window
• X-RateLimit-Reset — Unix timestamp when the window resets

When you exceed the limit the API returns 429 Too Many Requests. Back off and retry after X-RateLimit-Reset.

POST /v1/generate

The core endpoint. Creates a generation in the specified AI mode.

1curl https://api.forjegames.com/v1/generate \
2  -H "Authorization: Bearer $FORJEGAMES_API_KEY" \
3  -H "Content-Type: application/json" \
4  -d '{
5    "mode": "build",
6    "prompt": "A small medieval tavern with an NPC bartender",
7    "project_id": "proj_abc123"
8  }'

Request body:

• mode (string, required) — one of build, think, plan, image, script, terrain, 3d, debug, ideas
• prompt (string, required) — natural-language input
• project_id (string, optional) — attach to an existing project
• image_url (string, optional) — reference image for image mode
• stream (boolean, default false) — enable Server-Sent Events

Response:

1{
2  "id": "gen_01HE2Z9M5J0XJFKC7Q4YAF1DNX",
3  "mode": "build",
4  "project_id": "proj_abc123",
5  "status": "completed",
6  "credits_used": 28,
7  "output": {
8    "scripts": [
9      { "path": "ServerScriptService/Bartender.server.lua", "source": "..." }
10    ],
11    "parts": [ /* ... */ ],
12    "summary": "Built a 12x8 tavern with an NPC bartender script."
13  }
14}

Streaming

Pass "stream": true to receive output as Server-Sent Events. Each event is a JSON object with a type discriminator: thought, tool_call, file, done.

1const res = await fetch('https://api.forjegames.com/v1/generate', {
2  method: 'POST',
3  headers: { Authorization: `Bearer ${key}`, 'Content-Type': 'application/json' },
4  body: JSON.stringify({ mode: 'build', prompt, stream: true }),
5})
6// Parse SSE stream
7const reader = res.body!.getReader()
8// ... dispatch events as they arrive
9 

Projects

• GET /v1/projects — list projects
• GET /v1/projects/:id — fetch a project including chat history
• POST /v1/projects — create an empty project
• DELETE /v1/projects/:id — delete a project

Exports

• POST /v1/projects/:id/export — trigger a .rbxl export
• GET /v1/exports/:id — poll status and retrieve download URL

Errors

Errors follow a consistent shape:

1{
2  "error": {
3    "code": "insufficient_credits",
4    "message": "This action requires 28 credits but only 12 are available.",
5    "request_id": "req_01HE3A1NZ..."
6  }
7}

Common codes: invalid_api_key, insufficient_credits, rate_limited, invalid_mode, project_not_found, internal_error.