---

🦴 Slash Command

A Slash Command is a user-invokable workflow that starts with / and triggers predefined prompt sequences.

---

Key Characteristics

Property	Value
Invocation	🙋‍♀️ User types /command-name
Location	.claude/commands/*.md
Variables	$ARGUMENTS for user input
Execution	Synchronous, within main conversation

---

File Structure

# .claude/commands/generate.md

---
description: Generate localization files for specified locales
argument-hint: [locale]
---

Generate localization files for: $ARGUMENTS

1. First, identify if this is a single locale or language cluster
2. Check LOCALES-200.csv for valid locales
3. ...

Note: Command name comes from filename (generate.md → /generate).

---

Frontmatter Reference

Field	Required	Description
description	Recommended	Brief description (shown in /help)
argument-hint	No	Expected arguments hint
allowed-tools	No	Tools without permission prompts
model	No	Specific model to use
disable-model-invocation	No	Prevent programmatic invocation

---

Usage Examples

/generate fr-FR              # Single locale
/generate fr                 # Language cluster
/generate fr es de           # Multiple clusters

---

Mermaid Representation

%%{init: {'theme': 'base', 'themeVariables': {'lineColor': '#64748b'}}}%%
flowchart LR
    classDef user fill:#6366f1,stroke:#4f46e5,stroke-width:2px,color:#ffffff
    classDef main fill:#8b5cf6,stroke:#7c3aed,stroke-width:2px,color:#ffffff

    U["🙋‍♀️📥 /generate fr-FR"]:::user --> CMD["🦴 Slash Command"]:::user
    CMD --> MA["🐔💭 Main Agent"]:::main
    MA --> W["Workflow Execution"]
    W --> R["💁‍♀️📤 Result"]

---

Advanced: Nested Commands

Commands can reference other files in the .claude/commands/ directory:

# .claude/commands/full-review.md

---
description: Run comprehensive review pipeline
---

Execute the following reviews in sequence:
1. First run /security-review on $ARGUMENTS
2. Then run /performance-review on the same files
3. Finally run /style-check

---

Tips

• Keep commands focused on single concerns
• Use argument-hint for better discoverability
• Consider permission modes for destructive operations
• Document expected inputs and outputs

---