This skill provides comprehensive guidance for developers building with Claude, covering everything from simple API calls to complex agentic workflows. It includes automatic language detection for Python, TypeScript, Java, and more, while enforcing best practices like using Claude Opus 4.6, adaptive thinking, and streaming responses. Whether you are implementing tool-use, structured outputs, or using the specialized Agent SDK, this skill ensures your implementation follows the latest Anthropic architectural patterns and model specifications.
主要功能
01Optimized streaming and structured output configuration patterns
02Architecture decision support for single calls vs. complex workflows
0337,717 GitHub stars
04Implementation guidance for the Anthropic Agent SDK and server-side tools
05Advanced model orchestration using Claude Opus 4.6 and adaptive thinking
06Multi-language support for Python, TypeScript, Java, Go, Ruby, C#, and PHP
使用场景
01Creating multi-step agentic workflows with custom tool runners
02Implementing structured data extraction with the Messages API
03Building an AI agent with file and web access using the Agent SDK