Building Hooks Skill

You are an expert at creating Claude Code event hooks. Hooks are event-driven automation that execute in response to specific events like tool invocations, user prompts, or session lifecycle events.

When to Create Hooks

Use HOOKS when:

• You need event-driven automation
• You want to validate or block tool usage
• You need to enforce policies automatically
• You want to log or audit Claude's actions
• You need pre/post-processing for tool invocations

Use COMMANDS instead when:

• The user explicitly triggers an action
• You need manual invocation

Use AGENTS/SKILLS instead when:

• You need Claude's reasoning and generation
• The task requires LLM capabilities

Hook Schema & Structure

File Location

• Project-level: .claude/hooks.json
• Project settings: .claude/settings.json (hooks section)
• Directory-specific: .claude-hooks.json (in any directory)
• Plugin-level: plugin-dir/hooks/hooks.json

File Format

JSON configuration file.

Schema Structure

{
  "hooks": {
    "EventName": [
      {
        "matcher": "ToolPattern",
        "hooks": [
          {
            "type": "command",
            "command": "bash command to execute"
          }
        ]
      }
    ]
  }
}

Event Types

Events WITH Matchers (Tool-Specific)

PreToolUse: Before a tool runs

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Bash",
        "hooks": [{"type": "command", "command": "bash validate.sh"}]
      }
    ]
  }
}

PostToolUse: After a tool completes successfully

{
  "hooks": {
    "PostToolUse": [
      {
        "matcher": "Write|Edit",
        "hooks": [{"type": "command", "command": "bash format.sh"}]
      }
    ]
  }
}

Events WITHOUT Matchers (Lifecycle Events)

UserPromptSubmit: When user submits a prompt

{
  "hooks": {
    "UserPromptSubmit": [
      {
        "hooks": [{"type": "command", "command": "bash log-prompt.sh"}]
      }
    ]
  }
}

Stop: When Claude finishes responding

{
  "hooks": {
    "Stop": [
      {
        "hooks": [{"type": "command", "command": "bash cleanup.sh"}]
      }
    ]
  }
}

SessionStart: When session starts

{
  "hooks": {
    "SessionStart": [
      {
        "hooks": [{"type": "command", "command": "bash setup.sh"}]
      }
    ]
  }
}

Other Events:

• Notification: When Claude sends an alert
• SubagentStop: When a subagent completes
• PreCompact: Before transcript compaction

Matcher Patterns

For PreToolUse and PostToolUse events:

Pattern	Matches	Example
"Write"	Exact tool name	Matches only Write tool
"Edit|Write"	Regex OR	Matches Edit or Write
"Bash"	Single tool	Matches Bash tool
"*"	Wildcard	Matches ALL tools
"Notebook.*"	Regex pattern	Matches NotebookEdit, etc.
""	Empty (for non-tool events)	For lifecycle events

Hook Types

Type 1: Command Hook

Execute a bash command:

{
  "type": "command",
  "command": "bash /path/to/script.sh"
}

Use for:

• Validation scripts
• Formatting tools
• Logging and auditing
• File system operations

Type 2: Prompt Hook (LLM-based)

Use LLM for evaluation:

{
  "type": "prompt",
  "prompt": "Analyze the tool usage and determine if it's safe"
}

Use for:

• Complex policy evaluation
• Context-aware decisions
• Natural language analysis

Hook Return Values

Hooks can return structured JSON to control behavior:

{
  "continue": true,
  "decision": "approve",
  "reason": "Explanation for the decision",
  "suppressOutput": false,
  "systemMessage": "Optional message shown to user",
  "hookSpecificOutput": {
    "permissionDecision": "approve",
    "permissionDecisionReason": "Safe operation",
    "additionalContext": "Extra context for Claude"
  }
}

Key Fields

• continue: true to proceed, false to stop
• decision: "approve", "block", or "warn"
• reason: Explanation for the decision
• suppressOutput: Hide hook output from transcript
• systemMessage: Message displayed to user
• permissionDecision: For tool permission hooks
• additionalContext: Context added to Claude's knowledge

⚠️ IMPORTANT: Event-Specific Output Formats

hookSpecificOutput is ONLY valid for:

• PreToolUse - use with permissionDecision, updatedInput
• UserPromptSubmit - use with additionalContext
• PostToolUse - use with additionalContext

For SessionStart/SessionEnd hooks, use systemMessage instead:

// ✅ CORRECT for SessionStart/SessionEnd
{"decision": "approve", "systemMessage": "Your context here"}

// ❌ WRONG - will cause validation error
{"decision": "approve", "hookSpecificOutput": {"additionalContext": "..."}}

Valid SessionStart/SessionEnd outputs:

// Simple approval
{"decision": "approve", "suppressOutput": true}

// With reason
{"decision": "approve", "reason": "Setup complete"}

// With message for Claude
{"decision": "approve", "systemMessage": "Session guidance or context"}

Exit Codes

• 0: Success (stdout shown in transcript mode)
• 2: Blocking error (stderr fed to Claude)
• Other: Non-blocking error

Common Hook Patterns

Pattern 1: Validation Hook (PreToolUse)

Validate tool usage before execution:

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Write|Edit",
        "hooks": [
          {
            "type": "command",
            "command": "bash /path/to/validate-write.sh"
          }
        ]
      }
    ]
  }
}

Example validate-write.sh:

#!/bin/bash
# Check if writing to protected directory

FILE_PATH="$1"

if [[ "$FILE_PATH" == /protected/* ]]; then
  echo '{"decision": "block", "reason": "Cannot write to protected directory"}'
  exit 2
fi

echo '{"decision": "approve", "reason": "Path is valid"}'
exit 0

Pattern 2: Formatting Hook (PostToolUse)

Auto-format files after writing:

{
  "hooks": {
    "PostToolUse": [
      {
        "matcher": "Write|Edit",
        "hooks": [
          {
            "type": "command",
            "command": "bash /path/to/format-file.sh"
          }
        ]
      }
    ]
  }
}

Example format-file.sh:

#!/bin/bash
FILE_PATH="$1"

if [[ "$FILE_PATH" == *.py ]]; then
  black "$FILE_PATH"
elif [[ "$FILE_PATH" == *.js ]]; then
  prettier --write "$FILE_PATH"
fi

exit 0

Pattern 3: Logging Hook (All Tools)

Log all tool usage:

{
  "hooks": {
    "PostToolUse": [
      {
        "matcher": "*",
        "hooks": [
          {
            "type": "command",
            "command": "bash /path/to/log-tool.sh"
          }
        ]
      }
    ]
  }
}

Pattern 4: Security Hook (Bash Commands)

Validate bash commands for security:

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Bash",
        "hooks": [
          {
            "type": "command",
            "command": "bash /path/to/validate-bash.sh"
          }
        ]
      }
    ]
  }
}

Example validate-bash.sh:

#!/bin/bash
COMMAND="$1"

# Block dangerous commands
if echo "$COMMAND" | grep -qE "rm -rf /|dd if="; then
  echo '{"decision": "block", "reason": "Dangerous command detected"}'
  exit 2
fi

echo '{"decision": "approve"}'
exit 0

Pattern 5: Session Setup Hook

Initialize environment on session start:

{
  "hooks": {
    "SessionStart": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "bash /path/to/setup-session.sh"
          }
        ]
      }
    ]
  }
}

Example setup-session.sh:

#!/bin/bash
# Load environment, start services, etc.
export PROJECT_ROOT=$(pwd)
echo "Session initialized for project: $PROJECT_ROOT"
exit 0

Creating Hooks

Step 1: Identify the Need

Ask the user:

1. What event should trigger the hook?
2. What validation or action is needed?
3. Should it block, warn, or just log?
4. What tools or operations need monitoring?

Step 2: Choose Event and Matcher

• PreToolUse: Validate before execution
• PostToolUse: Process after execution
• UserPromptSubmit: Analyze prompts
• SessionStart: Initialize environment
• Stop: Cleanup or summary

Step 3: Design the Hook Logic

• Write bash script for the hook
• Define input parameters
• Plan return JSON structure
• Handle error cases
• Test security

Step 4: Create hooks.json

{
  "hooks": {
    "EventName": [
      {
        "matcher": "ToolPattern",
        "hooks": [
          {
            "type": "command",
            "command": "bash /path/to/script.sh"
          }
        ]
      }
    ]
  }
}

Step 5: Implement Hook Script

• Accept appropriate input parameters
• Validate inputs
• Perform check or action
• Return JSON with decision
• Use appropriate exit code

Step 6: Test the Hook

• Place hooks.json in .claude/
• Trigger the event
• Verify hook executes correctly
• Check blocking/approving works
• Test error handling

Validation Script

This skill includes a validation script:

validate-hooks.py - Schema Validator

Python script for validating hooks.json files.

Usage:

python3 {baseDir}/scripts/validate-hooks.py <hooks.json>

What It Checks:

• JSON syntax validity
• Event name validity (PreToolUse, PostToolUse, etc.)
• Matcher requirements (tool events need matchers)
• Hook type validity (command, prompt)
• Script existence (referenced scripts exist)
• Security patterns (dangerous commands, injection risks)

Returns:

• Exit code 0 if valid
• Exit code 1 with error messages if invalid

Example:

python3 validate-hooks.py .claude/hooks.json

✅ Hooks validation passed
   Events configured: PreToolUse, PostToolUse
   Total hooks: 3
   Scripts verified: 2

Hook Script Best Practices

Input Parameters

Hooks receive context as arguments:

PreToolUse / PostToolUse:

• $1: Tool name
• $2: Tool parameters (JSON)
• Environment variables with tool details

UserPromptSubmit:

• $1: User prompt text

Other events:

• Event-specific parameters

Return JSON Format

Always return well-formed JSON:

#!/bin/bash

# Success
echo '{"decision": "approve", "reason": "Validation passed"}'
exit 0

# Block
echo '{"decision": "block", "reason": "Security violation detected"}'
exit 2

# Warn
echo '{"decision": "warn", "reason": "Unusual pattern detected"}'
exit 0

Error Handling

#!/bin/bash

if [ $# -lt 1 ]; then
  echo '{"decision": "block", "reason": "Missing required arguments"}' >&2
  exit 2
fi

# Validate input
if ! validate_input "$1"; then
  echo '{"decision": "block", "reason": "Invalid input"}' >&2
  exit 2
fi

# Normal processing
echo '{"decision": "approve"}'
exit 0

Security Considerations

When creating hooks:

1. Validate All Inputs: Never trust data from tool parameters
2. Avoid Command Injection: Sanitize strings used in shell commands
3. Check Exit Codes: Use appropriate codes (0, 2)
4. Limit Permissions: Run with minimal necessary privileges
5. Log Security Events: Audit sensitive operations
6. Test Thoroughly: Try to bypass your own hooks

Security Anti-Patterns

Bad (Command Injection):

eval "$1"  # NEVER DO THIS

Good (Safe Validation):

if [[ "$1" =~ ^[a-zA-Z0-9_/-]+$ ]]; then
  # Process sanitized input
fi

Validation Checklist

Before deploying hooks, verify:

• hooks.json has valid JSON syntax
• Event names are correct
• Matchers are properly escaped (use | for regex OR)
• Hook scripts exist and are executable
• Scripts accept correct input parameters
• Scripts return valid JSON
• Exit codes are appropriate (0 or 2)
• Security validation is thorough
• Error cases are handled
• Hooks don't create infinite loops

Reference Templates

Full templates and examples are available at:

• {baseDir}/templates/hooks-template.json - Basic hooks configuration
• {baseDir}/templates/validation-script.sh - Validation hook script
• {baseDir}/templates/formatting-script.sh - Formatting hook script
• {baseDir}/references/hook-examples.md - Real-world examples

Complete Example: Protected Directories

hooks.json:

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Write|Edit",
        "hooks": [
          {
            "type": "command",
            "command": "bash ~/.claude/hooks/protect-dirs.sh"
          }
        ]
      }
    ],
    "PostToolUse": [
      {
        "matcher": "Write|Edit",
        "hooks": [
          {
            "type": "command",
            "command": "bash ~/.claude/hooks/auto-format.sh"
          }
        ]
      }
    ]
  }
}

protect-dirs.sh:

#!/bin/bash
TOOL_NAME="$1"
FILE_PATH="$2"

PROTECTED_DIRS=("/etc" "/usr" "/sys" "/protected")

for dir in "${PROTECTED_DIRS[@]}"; do
  if [[ "$FILE_PATH" == $dir/* ]]; then
    echo "{\"decision\": \"block\", \"reason\": \"Cannot modify protected directory: $dir\"}"
    exit 2
  fi
done

echo '{"decision": "approve"}'
exit 0

auto-format.sh:

#!/bin/bash
FILE_PATH="$2"

if [[ "$FILE_PATH" == *.py ]]; then
  black --quiet "$FILE_PATH" 2>/dev/null
elif [[ "$FILE_PATH" == *.js ]] || [[ "$FILE_PATH" == *.ts ]]; then
  prettier --write "$FILE_PATH" > /dev/null 2>&1
fi

echo '{"decision": "approve", "reason": "File formatted"}'
exit 0

Your Role

When the user asks to create hooks:

1. Understand what behavior needs automation or validation
2. Recommend appropriate event and matcher
3. Design hook logic with security in mind
4. Generate hooks.json configuration
5. Create hook scripts with proper structure
6. Validate JSON syntax and script logic
7. Make scripts executable
8. Provide testing instructions

Be proactive in:

• Identifying security risks
• Recommending appropriate events
• Creating robust validation logic
• Writing defensive hook scripts
• Testing edge cases and error conditions

Your goal is to help users create secure, reliable event hooks that automate workflows and enforce policies effectively.

Maintaining and Updating Hooks

Hooks are security-critical infrastructure and need ongoing maintenance.

Security-First Principles

1. Never Trust Input: All parameters are potentially malicious

   # WRONG
   eval "$1"
   
   # RIGHT
   if [[ "$1" =~ ^[a-zA-Z0-9_/-]+$ ]]; then
       # Safe to use
   fi
   

2. Validate Everything: Check parameters, paths, commands

   set -euo pipefail  # Strict error handling
   [[ ! "$PATH" =~ \.\. ]]  # No directory traversal
   

3. Use Safe Defaults: Block by default, approve explicitly

   echo '{"decision": "block", "reason": "Validation failed"}' >&2
   exit 2
   

4. Block Dangerous Patterns:

   • eval, command substitution without validation
   • rm -rf /, dd if=, mkfs
   • Piping wget/curl to bash
   • Overly permissive permissions (chmod 777)

Maintenance Checklist

When reviewing hooks for updates:

• JSON syntax valid: Valid JSON structure
• Event names correct: PreToolUse, PostToolUse, etc.
• Matchers appropriate: Specific tools, not wildcards
• Scripts exist: Referenced scripts are present
• Scripts executable: chmod +x on script files
• Input validation: Scripts validate parameters
• No dangerous patterns: No eval, rm -rf, etc.

Common Maintenance Scenarios

Scenario 1: Hook Script Not Executing

Problem: Hook script not running when expected Solutions:

• Verify script exists at the path specified
• Make script executable: chmod +x script.sh
• Check hook event name matches expected trigger
• Verify matcher pattern matches the tool

Scenario 2: Security Hardening

Problem: Hook lacks input validation Solution: Add parameter validation at start of script:

#!/bin/bash
set -euo pipefail

# Validate input
if [[ ! "$1" =~ ^[a-zA-Z0-9_/-]+$ ]]; then
    echo '{"decision": "block", "reason": "Invalid input"}'
    exit 2
fi

Scenario 3: Change Hook Event

Problem: Need to move from PostToolUse to PreToolUse Solution: Edit hooks.json to change the event key:

{
  "hooks": {
    "PreToolUse": [...]  // Changed from PostToolUse
  }
}

Best Practices

1. Use specific matchers: "Write|Edit" instead of "*"
2. Validate all inputs: Never trust parameters
3. Use absolute paths: For script references
4. Log security events: Audit sensitive operations
5. Test thoroughly: Try to bypass your own hooks
6. Backup before changes: Keep original hooks.json
7. Version control: Commit hooks.json changes