概要
This skill provides a comprehensive framework for running the Agentica SDK with a local Claude proxy, enabling Python-based agents to use the Claude CLI as their inference backend. It simplifies the complex orchestration of multiple local services, providing clear startup sequences, environment variable management, and robust health checks. Designed for developers building autonomous agents, it includes specific troubleshooting guides for SSE streaming issues, tool permission errors, and agent hallucinations, ensuring a stable and reliable development environment.
