/ Claude Skills / Claude Api
Development

Claude Api

"Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 β†’ 4.6, 4.6 β†’ 4.7, retired-model replacements). TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`; user asks for the Claude API, Anthropic SDK, or Managed Agents; user adds/modifies/tunes a Claude feature (caching, thinking, compaction, tool use, batch, files, citations, memory) or model (Opus/Sonnet/Haiku) in a file; questions about prompt caching / cache hit rate in an Anthropic SDK project. SKIP: file imports `openai`/other-provider SDK, filename like `*-openai.py`/`*-generic.py`, provider-neutral code, general programming/ML."

About this skill

# Building LLM-Powered Applications with Claude

This skill helps you build LLM-powered applications with Claude. Choose the right surface based on your needs, detect the project language, then read the relevant language-specific documentation.

Before You Start

Scan the target file (or, if no target file, the prompt and project) for non-Anthropic provider markers β€” import openai, from openai, langchain_openai, OpenAI(, gpt-4, gpt-5, file names like agent-openai.py or *-generic.py, or any explicit instruction to keep the code provider-neutral. If you find any, stop and tell the user that this skill produces Claude/Anthropic SDK code; ask whether they want to switch the file to Claude or want a non-Claude implementation. Do not edit a non-Anthropic file with Anthropic SDK calls.

Output Requirement

When the user asks you to add, modify, or implement a Claude feature, your code must call Claude through one of:

1. The official Anthropic SDK for the project's language (anthropic, @anthropic-ai/sdk, com.anthropic.*, etc.). This is the default whenever a supported SDK exists for the project.

2. Raw HTTP (curl, requests, fetch, httpx, etc.) β€” only when the user explicitly asks for cURL/REST/raw HTTP, the project is a shell/cURL project, or the language has no official SDK.

Never mix the two β€” don't reach for requests/fetch in a Pytho

…

Use cases

  • **Complexity** β€” Is the task multi-step and hard to fully specify in advance? (e.g., "turn this design doc into a PR" vs. "extract the title from this PDF")
  • **Value** β€” Does the outcome justify higher cost and latency?
  • **Viability** β€” Is Claude capable at this task type?
  • **Cost of error** β€” Can errors be caught and recovered from? (tests, review, rollback)
  • User asks for "latest" or "current" information
  • Cached data seems incorrect

Installation

  1. Download the ZIP using the button on the right.
  2. Unzip the folder.
  3. Drop the claude-api/ folder inside ~/.claude/skills/.
  4. Restart Claude Code (or your Claude Desktop session) β€” the skill loads automatically.
# macOS / Linux
mkdir -p ~/.claude/skills
unzip claude-skill-claude-api.zip -d ~/.claude/skills/

What's included

LICENSE.txt
SKILL.md
csharp/claude-api.md
go/claude-api.md
go/managed-agents/README.md
java/claude-api.md
java/managed-agents/README.md
php/claude-api.md
php/managed-agents/README.md
python/claude-api/README.md
python/claude-api/batches.md
python/claude-api/files-api.md
python/claude-api/streaming.md
python/claude-api/tool-use.md
python/managed-agents/README.md
ruby/claude-api.md
ruby/managed-agents/README.md
shared/agent-design.md
shared/error-codes.md
shared/live-sources.md
shared/managed-agents-api-reference.md
shared/managed-agents-client-patterns.md
shared/managed-agents-core.md
shared/managed-agents-environments.md
shared/managed-agents-events.md
shared/managed-agents-memory.md
shared/managed-agents-multiagent.md
shared/managed-agents-onboarding.md
shared/managed-agents-self-hosted-sandboxes.md
shared/managed-agents-tools.md
shared/managed-agents-webhooks.md
shared/model-migration.md
shared/models.md
shared/prompt-caching.md
shared/tool-use-concepts.md
typescript/claude-api/README.md
typescript/claude-api/batches.md
typescript/claude-api/files-api.md
typescript/claude-api/streaming.md
typescript/claude-api/tool-use.md
typescript/managed-agents/README.md

Frequently asked

What does the Claude Api skill do?
"Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 β†’ 4.6, 4.6 β†’ 4.7, retired-model replacements). TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`; user asks for the Claude API, Anthropic SDK, or Managed Agents; user adds/modifies/tunes a Claude feature (caching, thinking, compaction, tool use, batch, files, citations, memory) or model (Opus/Sonnet/Haiku) in a file; questions about prompt caching / cache hit rate in an Anthropic SDK project. SKIP: file imports `openai`/other-provider SDK, filename like `*-openai.py`/`*-generic.py`, provider-neutral code, general programming/ML."
Who should use it?
Anyone working with Claude on development-related tasks β€” consultants, founders, product teams, operators, and AI users who want repeatable, high-quality outputs from Claude.
How does it work?
Drop the folder into your `~/.claude/skills/` directory. Claude detects the SKILL.md instructions automatically and applies them when relevant.
What problem does it solve?
It removes the need to re-prompt Claude with the same context every time. Once the skill is installed, Claude already knows the rules, conventions, and procedures for this task.
How do I install it?
Download the ZIP, unzip it, and place the `claude-api/` folder inside `~/.claude/skills/`. No other setup required.
Is it free?
Yes β€” free download. We just ask for your name and work email so we can send useful Claude updates.
Can I get a custom version?
Yes. Abbas ElDeniney builds tailored Claude skills for specific business workflows and data sources. Book a consultation to scope your version.

Almost there

Share a few details and we'll start your download in one click.

Stay in the loop

Get notified the moment I publish a new blog, course, or case study.

πŸ”” Browser alerts
Instant pop-ups on this device