provider-integration

name	provider-integration
description	Guidelines for integrating video and audio generation providers. Includes API patterns, rate limits, and provider-specific gotchas.
context	fork
requires	{"tools":["Read","Edit","Write","Glob","Grep","Bash"],"model":"sonnet"}

Provider Integration Skill

Apply provider-specific patterns when implementing or debugging provider integrations.

When to Activate

• Adding a new video/audio/image provider
• Debugging provider API issues
• Optimizing provider usage for cost/quality
• Running provider onboard or provider test

Core Instructions

1. Check existing provider implementations in core/providers/
2. Follow the base provider interface pattern
3. Apply provider-specific guidelines from reference docs
4. Implement proper error handling and retries
5. Add to provider registry

Provider Interface

All providers implement:

class BaseProvider(ABC):
    @abstractmethod
    async def generate(self, request: GenerationRequest) -> GenerationResult:
        pass

    @abstractmethod
    def get_capabilities(self) -> ProviderCapabilities:
        pass

    @abstractmethod
    def estimate_cost(self, request: GenerationRequest) -> float:
        pass

Integration Checklist

• Implement base interface
• Add to PROVIDER_REGISTRY
• Handle rate limiting
• Implement cost estimation
• Add provider-specific tests
• Document in docs/providers.md
• Add memory learnings for gotchas

Return Value

When analyzing provider issues, return:

• provider_name: Which provider
• issue_type: API, rate_limit, quality, cost
• recommendation: Specific fix or workaround
• reference: Link to relevant guideline

Reference Documents

• luma-guidelines.md - Luma AI integration details
• elevenlabs-guidelines.md - ElevenLabs TTS details