What problem does it solve? Adding a new LLM provider to an Agno agent often leads to hardcoded model IDs, scattered os.getenv calls, and magic values without fallbacks. This Skill standardizes how providers are registered and configured so every model ID and constructor parameter comes from an environment variable with a pre-defined default. ## Core Features & Use Cases - Env-driven configuration: Enforces the rule that every model ID ({PROVIDER}_MODEL_ID) and every sampling parameter ({PROVIDER}_TEMPERATURE, _MAX_TOKENS, _TOP_P) is read from env with a default when absent or empty. - Provider adapters: Provides skeleton adapters for Google Gemini, xAI Grok, Moonshot Kimi, and AWS Bedrock with lazy Agno imports, LLMPort implementation, and composition_root wiring. - Typed env helpers: Supplies _get_str, _get_bool, and _get_opt_float helpers where empty values fall back to defaults and invalid values raise ValueError naming the env var. - Use Case: When adding Gemini support to an existing OpenAI/Groq agent, follow the checklist to register the provider, create the adapter, document GEMINI_MODEL_ID in .env.example, and resolve GOOGLE_API_KEY via the secrets pipeline. ## Quick Start Ask the agent to add a new Gemini provider adapter to the Agno project following the env-driven configuration rules with a default model ID.
