skills.rest
adrianasiam19
adrianasiam19COMMUNITY
Explore all skills by this author →
GitHub

claude-api

Build, debug, and migrate applications using the Claude API and Anthropic SDKs.

Updated Jan 31, 2026
Search Tags:#tool use#llm integration#claude api#prompt caching#model migration#anthropic sdk#managed agents
One-click install
npx skills add https://github.com/adrianasiam19/ATLAS --skill claude-api-adrianasiam19
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-api
Source: https://github.com/adrianasiam19/ATLAS/tree/main/.agent/skills/claude-api
Command: npx skills add https://github.com/adrianasiam19/ATLAS --skill claude-api-adrianasiam19

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Developers integrating Claude into their applications often struggle with correct SDK usage, model version migrations, prompt caching configuration, and choosing between single API calls, tool-use workflows, and Managed Agents. This Skill provides authoritative, language-specific guidance so generated code uses the correct Anthropic SDK bindings, current model IDs, and supported parameters. ## Core Features & Use Cases - Multi-language SDK guidance: Language detection routes to dedicated references for Python, TypeScript, Java, Go, Ruby, C#, PHP, and cURL, covering streaming, tool use, structured outputs, batch processing, and the Files API. - Model migration and feature configuration: Handles upgrades between Claude model versions (e.g., 4.5 to 4.6 to 4.7), adaptive thinking, effort tuning, compaction, and prompt caching with silent-invalidator diagnostics. - Managed Agents support: Walks through creating persistent, versioned server-managed agents with per-session workspaces, including an onboarding interview subcommand. - Use Case: A developer asks to add prompt caching to an existing Python Anthropic SDK project; the Skill detects the language, reads the caching reference, and produces code with correct cache_control breakpoints and usage verification. ## Quick Start Ask the assistant to add a streaming Claude API call with prompt caching to your Python or TypeScript file using the Anthropic SDK.

Frequently Asked Questions about claude-api

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

FAQPage Schema
How do I add prompt caching to a Claude API request?▼

Set cache_control with type ephemeral on the last block of stable content such as the system prompt, keeping volatile data after the final breakpoint. Verify hits by checking usage.cache_read_input_tokens; zero across repeated requests indicates a silent invalidator like timestamps in the prefix.

How do I migrate Claude API code to a newer model version?▼

Read the shared model-migration reference, which covers exact model IDs and breaking changes such as removed budget_tokens and prefill on Opus 4.6 and 4.7. Confirm the migration scope (file, directory, or codebase) with the user before editing.

Which languages does the Anthropic SDK support for tool use?▼

Beta tool runners exist for Python, TypeScript, Java, Go, Ruby, and PHP, each with language-specific patterns like decorators, Zod schemas, or annotated classes. C# has an official SDK without tool runner support, and other languages can use raw HTTP.

When should I use Managed Agents instead of the Claude API with tool use?▼

Use Managed Agents when you want Anthropic to run the agent loop and host a per-session container for bash, file operations, and code execution. It is first-party only and unavailable on Bedrock, Vertex AI, or Foundry, where Claude API plus tool use is required.

Why does my Claude API request return a 400 error with thinking enabled?▼

Opus 4.7 only accepts adaptive thinking; passing type enabled with budget_tokens returns a 400, as do sampling parameters like temperature. Use thinking type adaptive and control depth with the effort parameter inside output_config.

Related Skills

Explore More Software Engineering Skills→
el2060el2060

claude-api

Build and optimize applications with the Claude API and Anthropic SDK.

Community
Advanced
ndestatesndestates

token-usage-meter

Monitors xAI Grok session token usage and prompt-cache hit rates with cost estimates.

Community
Intermediate
CODE-SAURABHCODE-SAURABH
2

ai-engineer

Convert unreliable AI prototypes into production-ready systems with validation and observability.

Community
Advanced
pvliesdonkpvliesdonk
3

langchain-patterns

Provide reusable patterns for LangChain, LangGraph, and LiteLLM applications.

Community
Advanced