skills.rest
whatswithavi
whatswithaviCOMMUNITY
Explore all skills by this author →
GitHub

claude-api

Build LLM-powered applications with the Claude API and Anthropic SDKs.

Updated Mar 28, 2026
Search Tags:#agent sdk#streaming#tool use#llm integration#claude api#prompt caching#anthropic sdk
One-click install
npx skills add https://github.com/whatswithavi/Ecchelon_project --skill claude-api-whatswithavi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-api
Source: https://github.com/whatswithavi/Ecchelon_project/tree/main/.agent/skills/claude-api
Command: npx skills add https://github.com/whatswithavi/Ecchelon_project --skill claude-api-whatswithavi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Developers integrating Claude into their applications face fragmented documentation across multiple SDK languages, evolving model IDs, deprecated parameters, and subtle API behaviors like prompt caching invalidation and tool-use loops. This Skill provides language-specific, up-to-date guidance for building with the Claude API and Agent SDK without guesswork. ## Core Features & Use Cases - Multi-language SDK guidance: Language detection routes to tailored documentation for Python, TypeScript, Java, Go, Ruby, C#, PHP, and cURL, covering installation, streaming, tool use, thinking, and structured outputs. - Surface selection: Decision trees help choose between single API calls, tool-use workflows, and the Agent SDK based on task complexity, with current model IDs and pricing. - Feature deep-dives: Reference material covers prompt caching, compaction, batch processing, the Files API, server-side tools, and HTTP error handling. - Use Case: A developer building a TypeScript chat application with function calling gets routed to the TypeScript streaming and tool-use docs, uses the correct claude-opus-4-6 model ID, and implements the tool runner loop correctly on the first attempt. ## Quick Start Ask the assistant to help you build an application using the Claude API or Anthropic SDK in your project's programming language.

Frequently Asked Questions about claude-api

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

FAQPage Schema
How do I build an app with the Claude API?▼

Install the Anthropic SDK for your language, initialize a client with your API key, and call the Messages endpoint with a model ID, max_tokens, and messages array. The Skill detects your project language and provides tailored code examples for streaming, tool use, and error handling.

Which Claude model should I use for my application?▼

Default to claude-opus-4-6 for the highest capability, with claude-sonnet-4-6 and claude-haiku-4-5 as alternatives when the user explicitly requests them. Use exact model ID strings without date suffixes, and query the Models API for live capability lookups.

Claude API vs Agent SDK: which should I use?▼

Use the Claude API for single calls, code-orchestrated workflows, and custom agents with your own tools. Use the Agent SDK when Claude itself needs built-in file, web, or terminal access with permissions and MCP support; it is available for Python and TypeScript only.

Does the Anthropic SDK support my programming language?▼

Official SDKs exist for Python, TypeScript, Java, Go, Ruby, C#, and PHP, with cURL/raw HTTP as a fallback. Tool runner support is beta in most languages, while the Agent SDK is limited to Python and TypeScript.

Why is my prompt caching not working with the Claude API?▼

Cache hits fail when any byte in the prefix changes, such as timestamps in the system prompt or varying tool definitions. Keep stable content first, place cache_control breakpoints after it, and verify hits via usage.cache_read_input_tokens in the response.

How do I handle Claude API errors and rate limits?▼

Use the SDK's typed exception classes like RateLimitError and AuthenticationError instead of string-matching error messages. SDKs automatically retry 429 and 5xx errors with exponential backoff; check the retry-after header for rate limit recovery timing.

Related Skills

Explore More Software Engineering Skills→
unclehowellunclehowell

hermes-gateway-config

Configure Hermes gateway settings and manage API keys.

Community
Advanced
awsaws
2.3k

agents-optimize

Configure evaluators, observability, and cost governance for AgentCore agents.

Official
Advanced
hpsgdhpsgd

prompt-design

Design production prompts with structured frontmatter, versioning, and evaluation criteria.

Official
Advanced
reggiechan74reggiechan74
5

deployment

Configure Claude Code deployments on AWS Bedrock and Google Vertex AI.

Community
Advanced