skills.rest
leexb-wp21-prog
leexb-wp21-progCOMMUNITY
Explore all skills by this author →
GitHub

claude-api

Build Claude API integrations for streaming, tool use, and caching.

Updated Apr 28, 2026
Search Tags:#streaming#tool-use#claude-api#prompt-caching#model-migration#anthropic-sdk
One-click install
npx skills add https://github.com/leexb-wp21-prog/Azure-Ticket-Helpdesk --skill claude-api-leexb-wp21-prog
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-api
Source: https://github.com/leexb-wp21-prog/Azure-Ticket-Helpdesk/tree/main/.cursor/skills/skills-main/skills-main/skills/claude-api
Command: npx skills add https://github.com/leexb-wp21-prog/Azure-Ticket-Helpdesk --skill claude-api-leexb-wp21-prog

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes guesswork and boilerplate for building production-ready applications on the Claude (Anthropic) Messages API, including streaming, tool use, structured outputs, prompt caching, and model migration.

Core Features & Use Cases

  • Claude API app construction: Choose the right API surface (single call, tool-driven workflows, or agentic flows) and implement it safely.
  • Streaming for long outputs: Build real-time chat UIs and handle SSE/deltas without truncation issues.
  • Tool use and agents: Implement function calling, tool runners (where supported), and managed-agent steering patterns.
  • Prompt caching optimization: Add cache control breakpoints and troubleshoot low cache hit rates using usage metrics.
  • Model/version migration: Migrate between Claude model families and handle breaking parameter changes.
  • Prompt & output correctness: Use structured outputs (json schema) and validate results reliably.

Quick Start

Ask Claude API to generate a streaming, tool-enabled implementation for a TypeScript app that targets claude-opus-4-7 and uses prompt caching with verified cache hit rate checks.

Frequently Asked Questions about claude-api

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

FAQPage Schema
How do I implement streaming with the Claude API to handle long outputs without truncation?▼

Tool use in the Claude API lets you implement function calling and agent workflows by defining tools the model can invoke. You build tool runners to execute functions and manage agent steering patterns for complex, multi-step workflows.

How do I optimize prompt caching hit rates using the Anthropic SDK?▼

Model migration between Claude families requires handling breaking parameter changes safely when switching model versions. You update SDK-aligned API calls to accommodate new parameter structures and enforce correct handling for version-specific behaviors.

Can I use structured outputs with JSON schema in Claude API calls?▼

Structured outputs in the Claude API use JSON schema to enforce specific response formats and validate results reliably. You configure the schema within your API call to guarantee the model returns data matching your expected structure.

What is the best way to build a tool-enabled Claude app with prompt caching?▼

The best way to build a Claude app is combining SDK-aligned API calls for tool use with verified cache_control placement for prompt caching. You target specific models like claude-opus-4-7 while checking usage metrics to confirm cache hits.

Related Skills

Explore More Software Engineering Skills→
ComposioHQComposioHQ
29.6k

Building Agents using Anthropic with Composio

Integrate Anthropic's Claude API with Composio for tool orchestration.

Official
Advanced
rocklambrosrocklambros
2

profiling-llm-cost

Profile LLM application spend from trace logs with token counts and model pricing.

Community
Advanced
jezwebjezweb
961

claude-api

Integrate Claude API with JSON schema validation, streaming SSE, prompt caching, and tool calls.

Community
Intermediate
ObolNetworkObolNetwork
10

obol-stack-dev

Develop, test, and validate Obol Stack LLM smart-routing with llmspy gateway.

Official
Advanced