claude-api

Select Claude API or Agent SDK surfaces and provide language-specific implementation guidance.

1|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/pycode4micro/bushserver --skill claude-api-pycode4micro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-api
Source: https://github.com/pycode4micro/bushserver/tree/main/src/bushserver/seeds/skills/package/claude-api
Command: npx skills add https://github.com/pycode4micro/bushserver --skill claude-api-pycode4micro

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Helps you quickly implement Claude API or Anthropic SDK integrations by choosing the right surface (single call, tool use, or Agent SDK) and providing setup and architecture guidance tailored to your project language.

Core Features & Use Cases

  • Automatic integration guidance: Selects the appropriate approach (Claude API vs Agent SDK) based on whether you need files/web/terminal access, multi-step orchestration, or simple single-call tasks.
  • Language-aware setup: Detects the project language (Python/TypeScript/Java/Go/Ruby/cURL/C#/PHP) and routes you to the correct code examples and patterns.
  • Tooling, streaming, and reliability patterns: Covers tool use (tool runner/manual loop), streaming output, batching, files API, caching/compaction considerations, and structured outputs and error handling.

Quick Start

Tell an AI: "Implement a Claude API integration for my project in Python, using streaming and structured outputs, and add tool use for editing files when needed."

Frequently Asked Questions about claude-api

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I implement Claude API streaming and structured outputs in my application?

This Skill provides language-specific implementation guidance for Claude API streaming and structured outputs, automatically detecting your project language to supply the correct code examples and reliability patterns for your application.

When should I use the Anthropic Agent SDK instead of a single Claude API call?

You should use the Anthropic Agent SDK instead of a single Claude API call when your workflow requires multi-step orchestration, file/web/terminal access, or richer permissions. Single Claude API calls are better suited for simple Q&A and classification tasks.

Does the Claude API integration guide support TypeScript and Python for tool use?

Yes, the Claude API integration guide supports tool use for TypeScript and Python, alongside Java, Go, Ruby, cURL, C#, and PHP. It detects your project language and routes you to the correct tool runner and manual loop patterns.

What is the best way to handle errors and caching in Anthropic SDK multi-step workflows?

The best way to handle errors and caching in Anthropic SDK multi-step workflows is to apply the tooling, streaming, and reliability patterns provided by this Skill, which covers robust error handling guidance alongside caching and compaction considerations for multi-step orchestration.

Can I configure model thinking and effort parameters using the Claude API?

Yes, you can configure model thinking and effort parameters using the Claude API. This Skill provides implementation guidance that satisfies requirements for model and thinking/effort configuration alongside consistent tool-loop handling.