Summary &
Quick Reference

Tip: Use Haiku for routing/classification, Sonnet as your default, Opus for hard problems.

On macOS use Cmd, on Windows/Linux use Ctrl. Shortcuts may update — check Help menu.

Run claude --help for CLI flags: --model, --allowedTools, --print, --dangerously-skip-permissions

Permissions: tools require approval on first use. Grant with /permissions or allowedTools in settings.

# Installation
pip install anthropic

# Basic usage
import anthropic
client = anthropic.Anthropic()  # uses ANTHROPIC_API_KEY

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[
        {"role": "user",
         "content": "Explain quantum computing"}
    ]
)
print(message.content[0].text)

# Streaming
with client.messages.stream(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user",
               "content": "Write a poem"}]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

# With system prompt
message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    system="You are a helpful tutor.",
    messages=[{"role": "user",
               "content": "Explain gravity"}]
)

# Tool use
tools = [{"name": "get_weather", "description": "Get current weather",
  "input_schema": {"type": "object", "properties": {"location": {"type": "string"}}, "required": ["location"]}}]
msg = client.messages.create(model="claude-sonnet-4-20250514", max_tokens=1024, tools=tools,
  messages=[{"role": "user", "content": "Weather in Tokyo?"}])

// Installation
npm install @anthropic-ai/sdk

// Basic usage
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic(); // ANTHROPIC_API_KEY

const message = await client.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  messages: [
    { role: "user",
      content: "Explain quantum computing" }
  ],
});
console.log(message.content[0].text);

// Streaming
const stream = client.messages.stream({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  messages: [
    { role: "user", content: "Write a poem" }
  ],
});

for await (const event of stream) {
  if (event.type === "content_block_delta"
   && event.delta.type === "text_delta") {
    process.stdout.write(event.delta.text);
  }
}

const finalMessage = await stream.finalMessage();

// Multi-turn conversation
const messages = [
  { role: "user", content: "What is the capital of France?" },
  { role: "assistant", content: "The capital of France is Paris." },
  { role: "user", content: "What is its population?" }
];
const resp = await client.messages.create({ model: "claude-sonnet-4-20250514", max_tokens: 1024, messages });

# CLAUDE.md - Project Instructions for Claude Code

## Project Overview
This is a [type] project using [stack]. The main entry point is [file].

## Tech Stack
- Language: TypeScript 5.x / Python 3.12
- Framework: Next.js 14 / FastAPI
- Database: PostgreSQL with Prisma / SQLAlchemy
- Testing: Jest / pytest

## Key Commands
- `npm run dev` — start dev server
- `npm test` — run test suite
- `npm run lint` — lint and format

## Code Conventions
- Use functional components with hooks (React)
- Prefer named exports over default exports
- All functions must have docstrings/JSDoc
- Error handling: use Result types, never throw

## Architecture
- /src/components — React components
- /src/lib — shared utilities
- /src/api — API route handlers
- /src/db — database models and migrations

## Important Notes
- Never modify files in /src/generated/
- Always run tests before committing
- Use conventional commit messages

# agents.md - Instructions for Sub-Agents

## Role
You are a specialized agent focused on [specific task].
You have access to: Read, Grep, Glob tools only.

## Constraints
- Do NOT modify any files — only read and analyze
- Do NOT run shell commands
- Report findings in structured format
- Stay focused on the assigned subtask

## Output Format
Return your findings as:
1. Summary (1-2 sentences)
2. Relevant files (absolute paths)
3. Key findings (bulleted list)
4. Confidence level (high/medium/low)

## Context
The parent agent will provide you with a specific
question about the codebase. Answer it thoroughly
by searching and reading relevant files.

End of Series — Thank you for following along! Bookmark this reference deck.