Claude Code
Is a coding tool from Claude
Usage¶
- Work with an agentic loop
- Claude Copilot

- Claude code

- Claude Copilot
- Can manage it's own context window
- Recommended workflow: Explore → Plan → Code → Commit
- Explore & Plan in Plan mode
- Before Commit: run a code reviewer
- Ask
blah\to escape a line
Context¶
- Important to give Claude the right context / documents
- Claude Code
- Using
@file.ext
- Overview

- Escape to interrupt thinking,
continueto resume - Escape x2 to rewind memory
- Working memory
/contextto check in the Working Memory/compactmaintain knowledge but free up some context/clear
Commands¶
/init- Scan codebase
- Create summary
- Write in Claude Code
- Review
/commit-push-pragent
/goal/insightfor usage patterns/batchto fan out works/loopor/schedule!commandto run something (shell mode)/focusshow final result← ←to move from one CLI to available agents
Custom Commands¶
- In
.claude\commands\COMMAND-NAME.md- Put a prompt in it
- Use
$ARGUMENTSto pass args
- Restart Claude to refresh the command
Example that can be called with /write_tests the use-auth.ts file in the hooks directory
Write comprehensive tests for: $ARGUMENTS
Testing conventions:
* Use Vitest with React Testing Library
* Place test files in a __tests__ directory in the same folder as the source file
* Name test files as [filename].test.ts(x)
* Use @/ prefix for imports
Coverage:
* Test happy paths
* Test edge cases
* Test error states
Modes¶
- Plan mode
- Ask before edits
- Auto mode
- Read-only tools5
- Create a plan that can be executed
- Safe code review
- Complex changes
Tools¶
- OOTB tools
/effortto manage thinking- Keyword ultrathink
CLAUDE.MD¶
- Persistent memory
- At the Project and User
- CLAUDE.md
- CLAUDE.local.md
- ~/.claude/CLAUDE.md
- Reference project docs with
@file.ext - Start without one, then run
/initto generate it - Use
/memory somethingto add something to CLAUDE.md - Use
@file.extto mention a specific file
Capabilities¶
MCP¶
/mcp- Scope
- Local
- User
- Project
- Add into context window
SDK¶
npm install @anthropic-ai/claude-agent-sdk- See documentation
- Enable Claude to be used programmatically
- Read-only access by default
- Inherit project if launched inside claude

