How do I implement Prompt Caching to save on costs?

The skill provides patterns for using the 'cache_control' parameter with 'ephemeral' types in system prompts and message blocks, potentially saving up to 90% on input costs.

Which Claude models are supported by this skill?

This skill supports the latest Anthropic models including Claude Opus 4.5, Sonnet 4, 3.5 Sonnet, and 3.5 Haiku, providing context limits and pricing data for each.

Does this skill provide support for TypeScript developers?

Yes, it includes full implementation examples for both the @anthropic-ai/sdk in TypeScript and the anthropic package in Python.

What is the 'Extended Thinking' feature?

Extended Thinking allows Claude 4 models to use a specific token budget for deep reasoning before providing a final response, ideal for complex coding or math problems.

How can I handle API rate limits and connection errors?

The skill includes specialized error handling logic for AuthenticationError, RateLimitError, and APIConnectionError with exponential backoff examples.

Anthropic Platform Integration

Name: Anthropic Platform Integration
Author: FortiumPartners

byFortiumPartners

0•

API Development

Facilitates seamless integration with the Anthropic Claude SDK for implementing Messages API, tool use, and advanced reasoning features.

This skill provides a comprehensive toolkit for developers building with the Anthropic platform, offering domain-specific guidance on the Messages API, streaming, and tool use. It helps developers select the right Claude models—ranging from Opus to Haiku—while implementing advanced capabilities like Extended Thinking for complex reasoning, Vision for multimodal inputs, and Prompt Caching for significant cost reduction. By providing ready-to-use patterns for Python and TypeScript, this skill ensures robust error handling and efficient AI-augmented development workflows within the Claude Code environment.

Key Features

01Configuration support for Extended Thinking and thinking budgets

02Optimized strategies for Prompt Caching and Message Batches to reduce costs

03Ready-to-use code snippets for Python and TypeScript SDKs

04Comprehensive guidance for Messages API and multi-turn conversations

05Implementation patterns for Tool Use and function calling logic

060 GitHub stars

Use Cases

01Implementing vision-enabled applications using base64 and PDF support

02Building sophisticated AI agents with tool-calling capabilities

03Optimizing production AI workflows for cost and performance using caching

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add fortiumpartners/ensemble using-anthropic-platform

For use in Claude.ai and ChatGPT

Key Features

01Configuration support for Extended Thinking and thinking budgets

02Optimized strategies for Prompt Caching and Message Batches to reduce costs

03Ready-to-use code snippets for Python and TypeScript SDKs

04Comprehensive guidance for Messages API and multi-turn conversations

05Implementation patterns for Tool Use and function calling logic

060 GitHub stars

Use Cases

01Implementing vision-enabled applications using base64 and PDF support

02Building sophisticated AI agents with tool-calling capabilities

03Optimizing production AI workflows for cost and performance using caching

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add fortiumpartners/ensemble using-anthropic-platform

For use in Claude.ai and ChatGPT