Implement Claude API integrations with provider-specific SDK usage and caching.
npx skills add https://github.com/lth0/codexSkill --skill claude-api-lth0
Please help me install this Agent Skill. Skill: claude-api Source: https://github.com/lth0/codexSkill/tree/main/skills/agents/claude-api Command: npx skills add https://github.com/lth0/codexSkill --skill claude-api-lth0
This Skill helps you design and implement Claude (Anthropic) API integrations that work correctly across common surfaces (single-call, streaming, tool use, and managed agents) while enforcing best practices like provider consistency, prompt caching, and robust tool-result handling.
POST /v1/messages correctly and choose the right request surface for your workflow.Use the claude-api skill when you want to add Claude API support to an app by using the official Anthropic SDK for your language, enabling prompt caching where appropriate, and implementing tool-use handling (streaming and tool-result round-trips) if your feature requires function calls.
High-intent search queries and answers about installing and using this skill.
To implement tool use with the Claude API, you need to construct function calling patterns using tool runners or manual loops, ensuring safe tool-result continuation logic without mixing incompatible SDK providers. This skill provides production-ready patterns for agentic loops and streaming tool round-trips.
Prompt caching optimization in the Anthropic API requires maintaining prefix-stability rules, placing cache breakpoints correctly, and verifying cache hits. This skill guides you through caching verification to ensure your application correctly leverages prompt caching for reduced latency.
Streaming with the Claude API works by handling long responses and real-time interfaces through the official Anthropic SDK. This skill provides streaming handling guidance to build UI-ready output that processes continuous data flows correctly across single-call and agentic workflows.
Yes, you can use Managed Agents beta workflows with the Claude API by following the agent to session flow, event streaming, and steering integration constraints. This skill explains how to implement these beta features while maintaining provider-specific SDK usage and adaptive thinking defaults.
Claude API applications fail when mixing incompatible providers or SDK abstractions because the API requires strict provider consistency and correct model ID usage. This skill enforces provider-specific SDK usage and safe tool-result continuation to prevent integration failures.
Yes, building Claude API apps correctly requires adaptive thinking defaults, particularly for Claude 4.6 models, along with effort tuning. This skill recommends adaptive thinking configurations and explains how to satisfy strict requirements for model ID usage and API surface selection.
wenermeProvides Ruby SDK method and endpoint guidance for Anthropic API workflows.
anilveersingh1308Implement Claude SDK applications with Messages API, tool use, and streaming.
ZateDetect language-specific vulnerabilities in JavaScript, Python, Go, Java, Ruby, and PHP code.
tuyenhtDebug Vercel AI SDK v5 streaming, tool calling, and provider integration issues.