Introduction
The Credo AI SDK provides client libraries for interacting with the Credo AI platform. Available for Python and TypeScript, it enables you to programmatically manage AI governance workflows, including use cases, models, vendors, and their relationships.
What You Can Do
With the SDK, you can:
- Manage Use Cases - Create, update, and organize AI/ML applications for governance tracking
- Register Models - Track AI/ML models and their metadata across your organization
- Track Vendors - Monitor third-party AI providers and tools in your ecosystem
- Define Relationships - Link models and vendors to use cases for comprehensive governance
Key Features
| Feature | Python | TypeScript |
|---|---|---|
| Client | Sync CredoAI and async AsyncCredoAI | Single async client via createCredoAIClient() |
| Type Safety | Pydantic models for request/response validation | Full TypeScript types for all resources |
| Pagination | Built-in list_all() generators | Cursor-based manual pagination |
| Error Handling | Exception-based (ApiError, AuthenticationError) | Result-based { data, error } pattern |
Resources
The SDK provides access to these resources:
| Resource | Python | TypeScript |
|---|---|---|
| Use case management | client.use_cases | client.useCases |
| Model registry | client.models | client.models |
| Vendor tracking | client.vendors | client.vendors |
| Use case → model relationships | client.use_case_models | client.useCases.models |
| Use case → vendor relationships | client.use_case_vendors | client.useCases.vendors |
| Model → vendor relationships | client.model_vendors | client.models.vendors |
| Health checks and metrics | client.system | client.system |