Claude API
AI & MLby anthropics
2/3 audits passApache-2.0
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file;
Claude API, from Anthropic's skills repository, is a reference for building applications on the Claude API and Anthropic SDKs. Before editing, it tells the agent to scan the target code for another provider's markers and stop to ask rather than mixing providers. Code must call Claude through the official SDK for the project's language, or raw HTTP only when appropriate, and the agent must never guess SDK usage: bindings come from the skill's per-language files or official sources. It covers choosing an API surface and whether to build an agent, current models, authentication, thinking and effort, compaction, prompt caching, task budgets, cloud provider clients, context editing, Managed Agents, server tools, file input and tool-use patterns, with language folders from Python and TypeScript to Go, Java, C#, PHP and Ruby.
It also defines subcommands: migrate to a newer model, prompt-audit for dated prompting patterns, upgrade for a major SDK version, and cost-optimize to reduce spend. It is not for code built on other LLM providers.
This is a skill for coding agents such as Claude Code. It is published under Apache-2.0 and importable into BusinessMCP, though its use is in application code.
What you can do with it
- Add a Claude call using the correct official SDK
- Migrate existing code to a newer Claude model
- Audit prompts and tool descriptions for dated patterns
- Find cost savings through caching and batching
Run it on your business data
Imported into BusinessMCP, Claude API becomes a playbook your AI business analyst applies to your connected tools and MCP servers.
Use Claude API in BusinessMCPInstall it in a coding agent
One command adds Claude API to your project.
npx skills add https://github.com/anthropics/skills --skill claude-apiHow we vetted it
- Source
- anthropics/skills at 3337550
- Licence
- Apache-2.0
- Security audits (skills.sh)
- Gen Agent Trust Hub: Pass · Socket: Pass · Snyk: Warn
- Bundled scripts
- None, instructions only
- Reviewed by hand
- documents downloading a binary or archive; includes commands that use sudo.
Checked 2026-09-25 against its skills.sh listing. How we vet skills
Related skills
All skillsMCP Builder
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools.
Skill Creator
Create new skills, modify and improve existing skills, and measure skill performance.
Hugging Face LLM Trainer
Train or fine-tune language and vision models using TRL (Transformer Reinforcement Learning) or Unsloth with Hugging Face Jobs infrastructure.
Frequently asked questions
Will Claude API edit code that uses another LLM provider?
No. If it finds another provider's markers it stops and asks whether to switch the file to Claude or keep a non-Claude implementation.
Is Claude API safe to install?
Our static scan flagged binary-download and sudo because reference files document downloading the ant CLI into /usr/local/bin with sudo and creating a memory directory with sudo for self-hosted sandboxes. These are documented setup steps, and the skill passed the skills.sh security auditors with no FAIL.
What subcommands does it offer?
migrate, prompt-audit, upgrade and cost-optimize, each of which reads a bundled guide and follows it step by step.