Overview
Apollo’s backend is a Python FastAPI application that provides REST API endpoints for all platform features. The API server runs at:- Production:
api.apol.dev - Development:
dev-api.apol.dev
API Architecture
All endpoints require authentication via Supabase JWT tokens passed in theAuthorization: Bearer <token> header.
Router Summary
| Router | Prefix | Description |
|---|---|---|
| AI | /ai | AI features — chat, NLP, PSE generation |
| Apollo Agent | /api/v1/agent | Apollo Brain chat, streaming, sessions |
| Voice | /api/v1/voice | OpenAI Realtime API token provisioning |
| Audio | /audio | Whisper transcription |
| Knowledge Base | /kb | KB pages and folders CRUD |
| GitHub | /github | GitHub read operations |
| GitHub OAuth | /github-oauth | GitHub OAuth flow |
| GitHub Webhooks | /webhooks/github | Incoming GitHub events |
| Cloud | /cloud | AWS account/resource/cost management |
| Infrastructure | /infrastructure | ECS/Lambda/ECR/Amplify/Terraform |
| Deployments | /deployments | Amplify deployment management |
| PSE | /pse | Project Spec Engine sessions |
| AI Workflow | /ai-workflow | AI Workforce sessions/jobs |
| Content Engine | /content | Content creation and management |
| Automations | /automations | Automation CRUD and execution |
| Google Calendar | /google-calendar | Calendar events and sync |
| Organizations | /organizations | Organization management |
| Admin | /admin | Admin operations |
| Email Builder | /email | Email template management |
| Platform Engineering | /platform-engineering | Unified provider resources |
Common Patterns
Error Responses
All endpoints return standard error responses:Pagination
List endpoints support pagination viaoffset and limit query parameters.
Rate Limiting
- Standard endpoints: 100 requests/minute per user
- Voice endpoints: 20 requests/hour, 50/day per user
- AI endpoints: Rate-limited based on model costs