claude-api

Integrate Claude API and Agent SDK across seven programming languages.

Updated Mar 31, 2026
One-click install
npx skills add https://github.com/DFly7/iOS-FastAPI-Supabase-AI --skill claude-api-dfly7
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-api
Source: https://github.com/DFly7/iOS-FastAPI-Supabase-AI/tree/main/.agents/skills/claude-api
Command: npx skills add https://github.com/DFly7/iOS-FastAPI-Supabase-AI --skill claude-api-dfly7

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a single, opinionated reference for integrating Anthropic Claude models and the Agent SDK into applications so teams avoid trial-and-error across languages, streaming, tool use, and agent workflows.

Core Features & Use Cases

  • Language detection and curated, language-specific examples for Python, TypeScript, Java, Go, Ruby, C#, and PHP.
  • Defaults and best practices for model selection, adaptive thinking, streaming, and long-output handling.
  • Tooling patterns: beta tool runners, manual agentic loops, server-side tools (code execution, web search, web fetch), and Files API usage.
  • Prompt caching, compaction, structured outputs, and error-handling guidance to build robust, production-grade integrations.
  • Examples and patterns for batch processing, token counting, and MCP / Agent SDK usage for agents that need file/web/terminal access.

Quick Start

Ask for language-specific Claude API examples, for example: "Show me Python examples to call the Claude API with streaming, adaptive thinking, tool use, and prompt caching."

Frequently Asked Questions about claude-api

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

FAQPage Schema
How do I integrate Claude API streaming responses in Python?

To integrate Claude API streaming in Python, you can request language-specific examples that cover streaming patterns, adaptive thinking, and tool use. The Skill provides curated Python snippets for production-grade Claude integrations with proper response handling.

What is the best way to build an agentic loop with the Anthropic Agent SDK?

The best way to build an agentic loop with the Anthropic Agent SDK is to use the manual agentic loop patterns provided. These patterns demonstrate tool-runner implementations and server-side tool execution for robust agent workflows.

Does the Claude API support prompt caching and file uploads?

Yes, the Claude API supports prompt caching and file uploads through the Files API. The Skill provides best practices for implementing both features alongside structured outputs and token counting for optimized API usage.

Can I use Claude API examples for Go, Ruby, and PHP applications?

Yes, you can use Claude API examples for Go, Ruby, and PHP applications. The Skill detects your language and provides curated, language-specific implementation examples for Python, TypeScript, Java, Go, Ruby, C#, and PHP.

How do I handle long-output and compaction when calling Claude models?

To handle long-output and compaction when calling Claude models, apply the defaults and best practices for long-output handling and compaction provided. The Skill includes error-handling guidance to ensure robust production-grade integrations.

What model defaults should I use for Claude API production integrations?

For Claude API production integrations, you should use model defaults like claude-opus-4-6. The Skill specifies requirements for model selection, adaptive thinking, and structured outputs to ensure reliable application behavior.