Skip to main content
Service Name: integration-api
Port: 9004
Technology: Go + Gin Framework
Language: Go 1.25+
Primary Database: PostgreSQL (integration_db)

​
Purpose

The Integration API handles:
  • LLM provider integration (OpenAI, Anthropic, etc.)
  • STT provider integration (Google, Deepgram, etc.)
  • TTS provider integration (ElevenLabs, Azure, etc.)
  • Credential encryption and secure storage
  • Provider health monitoring
  • OAuth 2.0 authentication flows
  • Provider model listing and validation
  • API rate limiting and cost tracking

​
Key Features

​
Provider Management

  • 20+ pre-configured providers
  • Custom provider support
  • Provider-specific configuration
  • Model listing and caching
  • Provider health monitoring

​
Credential Security

  • AES-256-GCM encryption
  • Secure key management
  • Per-credential encryption
  • Audit logging of access
  • Credential rotation support

​
OAuth Integration

  • Multiple OAuth providers (Google, GitHub, Slack, etc.)
  • Token refresh handling
  • Scope management
  • Secure token storage
  • Revocation support

​
API Validation

  • Test API credentials before storage
  • Provider-specific validation
  • Rate limit tracking
  • Cost estimation
  • Usage monitoring

​
Configuration

​
Environment Variables

# Service
SERVICE_NAME=integration-api
PORT=9004
HOST=0.0.0.0
ENV=production
LOG_LEVEL=info

# Database
POSTGRES__HOST=postgres
POSTGRES__PORT=5432
POSTGRES__DB_NAME=integration_db
POSTGRES__AUTH__USER=rapida_user
POSTGRES__AUTH__PASSWORD=rapida_db_password
POSTGRES__MAX_OPEN_CONNECTION=15
POSTGRES__MAX_IDEAL_CONNECTION=8
POSTGRES__SSL_MODE=disable

# Encryption
INTEGRATION_CRYPTO_KEY=base64-encoded-256-bit-key
CRYPTO_ALGORITHM=AES-256-GCM

# Redis (caching)
REDIS__HOST=redis
REDIS__PORT=6379
REDIS__DB=1
REDIS__MAX_CONNECTION=10

# OAuth Configuration
OAUTH_CALLBACK_URL=http://localhost:8080/oauth/callback
GOOGLE_OAUTH_CLIENT_ID=
GOOGLE_OAUTH_CLIENT_SECRET=
GITHUB_OAUTH_CLIENT_ID=
GITHUB_OAUTH_CLIENT_SECRET=

# Provider-specific (optional defaults)
DEFAULT_OPENAI_MODEL=gpt-4
DEFAULT_STT_PROVIDER=google
DEFAULT_TTS_PROVIDER=google

​
Source Code Structure

api/integration-api/
├── api/                    # REST handlers
│   ├── integration.go      # Integration endpoints
│   ├── openai.go           # Provider-specific handlers
│   ├── anthropic.go
│   ├── cohere.go
│   ├── azure.go
│   ├── gemini.go
│   ├── embedding.go        # Embedding models
│   ├── reranking.go        # Reranking models
│   └── health/
│
├── internal/
│   ├── entity/             # Data models
│   │   └── external_audit.go
│   │
│   ├── service/            # Business logic
│   │   ├── audit.service.go
│   │   └── audit/
│   │
│   ├── caller/             # Provider API callers
│   │   ├── callers.go
│   │   ├── anthropic/
│   │   │   ├── anthropic.go
│   │   │   ├── llm.go
│   │   │   └── verify-credential.go
│   │   ├── openai/
│   │   │   ├── openai.go
│   │   │   ├── llm.go
│   │   │   ├── embedding.go
│   │   │   └── verify-credential.go
│   │   ├── azure/
│   │   ├── cohere/
│   │   ├── gemini/
│   │   ├── mistral/
│   │   ├── huggingface/
│   │   ├── vertexai/
│   │   ├── replicate/
│   │   ├── voyageai/
│   │   └── metrics/
│   │       └── metrics_builder.go
│   │
│   └── store/              # Database access
│
├── migrations/             # Database migrations
├── router/                 # Route definitions
├── config/
└── main.go

​
Credential Test Fails

  • Verify API key is correct
  • Check provider status page
  • Verify API key has required permissions
  • Check rate limiting hasn’t kicked in

​
Next Steps