Claude Code Reference

v2.1.178 · 87 auto-discovered · Commands, shortcuts, hooks, Python/AWS tips.

Slash Commands

Click a command to see its options and example invocations.

Anthropic’s Claude Mythos PreviewAUTO
A gated defense framework of Project Glasswing that represents a massive leap in AI capability for cybersecurity.N/A
Model Context Protocol (MCP) supportAUTO
Enables AI assistants like Claude Code to interact with Terraform, reducing manual effort and accelerating delivery.N/A
Claude CodeAUTO
handles many day-to-day engineering tasks effectivelyN/A
Cursor and Claude CodeAUTO
handles many day-to-day engineering tasks effectivelynot applicable
skills fileAUTO
teach the agent about blog's architecture, deployment, and what not to touchN/A
mass-migrated blog to AstroAUTO
Migrate a large number of posts and assets to Astro using Claude Code.Not applicable
mass-migrating blog to AstroAUTO
Use Claude Code to migrate a large number of blog posts to Astro.Not specified
Isolation and access controlAUTO
Strictly controlling access to private backends and customizing agent's toolsNot specified
Deployment and architecture understandingAUTO
Teaching the agent about blog's architecture and deploymentNot specified
Astro migration supportAUTO
Mass-migrate a blog to Astro using Claude CodeNot specified
Customizing agent's toolsAUTO
easily customize agent's toolsnot specified
Autonomous code deliveryAUTO
scale agent workflows globally while controlling access to private backendsnot specified
IsolationAUTO
teach the agent about blog's architecture and what not to touchnot specified
Sitemap generationAUTO
generate sitemaps using Claude Codenot specified
JSON-LD markupAUTO
generate JSON-LD markup using Claude Codenot specified
DeploymentAUTO
handle deployment of blog using Claude Codenot specified
Astro migrationAUTO
mass-migrate a blog to Astro using Claude Codenot specified
ObservabilityAUTO
Get observability in the terminal for you and your agents with the gcx CLI toolNot specified
Claude Managed AgentsAUTO
Cloudflare integration for fast, isolated execution environment for autonomous code deliveryNot specified
Skills fileAUTO
Teach the agent about blog architecture, deployment, and what not to touchNot specified
Cloudflare integrationAUTO
Cloudflare has integrated with Anthropic's Claude Managed Agents for autonomous code deliveryNot applicable
Isolation and deploymentAUTO
spent hours writing a skills file to teach the agent about blog's architecture, deployment, and what not to touchNot applicable
teaching agent about blog architectureAUTO
Write a skills file to teach Claude Code about your blog's architecture and deployment.Not provided
Specialized Deep Research SkillAUTO
Adds deep research capabilities to agent harnesses like Claude CodeN/A
Agentic toolsAUTO
Handle many day-to-day engineering tasksNot applicable
DeepSeek V4 ProAUTO
An enhancement to the Claude Code agent loop for improved performanceNot specified
Commit PolicyAUTO
Refuses requests or charges extra if commits mention 'OpenClaw'N/A
Commit filteringAUTO
Refuses requests or charges extra if commits mention 'OpenClaw'N/A
/help
Show help menu
/config
Open settings interface
/model
Switch the active model mid-session.

Accepted values

  • claude-opus-4-7Latest flagship — best for hard agentic coding.
  • claude-opus-4-6Previous flagship, still very capable.
  • claude-sonnet-4-6Fast & cheap default.
  • claude-haiku-4-5Fastest/cheapest — small tasks.

Examples

/model claude-opus-4-7
/model claude-sonnet-4-6

TipOr use Cmd+P / Meta+P to open the picker.

/effort
Set extended-thinking effort for the rest of the session.

Accepted values

  • lowSnappy, minimal thinking.
  • mediumBalanced default.
  • highMore reasoning — tricky bugs.
  • maxMaximum depth — hardest problems.

Examples

/effort high
/effort max
/clear
Clear conversation and start fresh
/compact
Manually compact context window
/resume
Resume previous sessions (open picker)
/rename
Rename current session
/plan
View current plan (in plan mode)
/usage
View token usage and costs
/version
Show Claude Code version
/ide
Connect to VS Code or JetBrains IDE
/thinking-toggle
Toggle extended thinking on/off
/agents
View and create custom subagents
/skills
Browse available skills
/plugins
Browse and manage plugins
/mcp
Add and configure MCP servers
/hooks
Browse configured hooks
/permissions
View and manage tool permissions
/memory
View project memory (CLAUDE.md)
/debug
Enable debug logging for troubleshooting
/simplify
Review code for reuse, quality, and fix issues
/batch
Plan and execute large-scale changes across 5-30 isolated worktree agents
/loop
Run a prompt or command on a recurring interval
/schedule
Create scheduled remote agents on cron schedule
/claude-api
Build, debug, and optimize Claude API / Anthropic SDK apps
/find-docs
Retrieve up-to-date docs for any library (via Context7)
/init
Initialize CLAUDE.md for your project
/review
Review a pull request
/security-review
Perform security review of pending changes
/insights
Generate report analyzing Claude Code sessions
/install-github-app
Install Claude GitHub App for CI/CD
/powerup
Interactive lessons with animated demos
/remote-control
Start Remote Control session