adapter-factory

What problem does it solve?

Integrating diverse AI models (CLI tools, HTTP APIs) into a unified deliberation system can be complex and error-prone. This skill provides a factory pattern and step-by-step guide to create robust adapters, enabling seamless participation of any AI model in your AI Counsel.

Core Features & Use Cases

• CLI Adapter Creation: Build adapters for command-line AI tools, handling subprocess execution and output parsing.
• HTTP Adapter Creation: Integrate HTTP-based AI APIs, managing requests, retries, and response parsing.
• Config-Driven Integration: Configure new adapters via YAML, supporting environment variable substitution for secrets.
• Test-Driven Development: Guides on writing unit and integration tests, including VCR for HTTP response recording.
• Use Case: You want to add support for a new local LLM (e.g., a custom fine-tuned model via Ollama) or a proprietary cloud API. This skill walks you through creating the necessary adapter, configuring it, and testing its integration into the deliberation system.

Quick Start

Show me how to create a new HTTP adapter for a custom AI API, including how to build the request and parse the response.