Claude API
82TrustedInstructions onlyTechnical implementation reference for the Anthropic model ecosystem.
What it does
This skill provides a structured knowledge base for integrating Anthropic models into software projects. It ensures that your assistant generates accurate, up-to-date code using the official SDKs, while avoiding common pitfalls associated with evolving API parameters and model-specific requirements.
By installing this, you give your assistant a reliable guide for handling authentication, streaming, tool definitions, and model-specific features like adaptive thinking. It prevents common errors caused by outdated API patterns and ensures consistency across different programming languages and project environments.
When it earns its place
Reach for it when…
Apply this when building or maintaining applications that interface with Anthropic models, especially when you need to switch between specific Claude versions, implement complex tool-use workflows, or optimize streaming responses for high-latency tasks.
Look elsewhere when…
Do not use this for projects explicitly anchored to other LLM providers like OpenAI, Google Gemini, or local models. It is also unnecessary for simple, non-programmatic text generation tasks that do not involve integrating or configuring the Anthropic API within a codebase.
What to say to use it
You never invoke a skill directly — the assistant picks it up when your request matches what it covers. These are the kinds of things that do that.
- How do I set up Claude in my Python script
- Switch my current prompt implementation to use Claude Opus 5
- Add streaming support to my existing Anthropic API calls
- What are the latest best practices for model thinking parameters
Install it
Install for all your projects
Copy the skill directory into your personal skills folder. Claude picks it up on its own the next time your request matches what it does — there is nothing to enable.
git clone --depth 1 --branch main https://github.com/anthropics/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/skills/claude-api ~/.claude/skills/claude-apiInstall for one project only
Put it in the project's own skills folder instead, so it can be committed and shared with everyone working on that repository.
cp -r /tmp/skills/skills/claude-api .claude/skills/claude-apiIf a skill of the same name exists in both places, the personal one wins.
Installing on Claude.ai or through the API instead? The full guide covers every surface.
Pair it with an MCP server
Claude API changes how the work is done. These servers change what Claude can reach while doing it.
- GitHub MCP Server53The skill sets the conventions the code should follow; GitHub MCP Server is what lets Claude read the repository it is writing into.
- Context Forge MCP Server86Context Forge MCP Server gives the code somewhere to actually run, so the skill's guidance can be checked instead of assumed.
- Supabase MCP Server100Pair with Supabase MCP Server when the code being written has to match a real schema rather than a described one.