Claude Code Hook Development FAQs

Question 1

How do I use the ${CLAUDE_PLUGIN_ROOT} variable?

Accepted Answer

This environment variable should be used in your hook commands to ensure portability, allowing scripts to correctly reference paths relative to the plugin's installation directory.

Question 2

How do I persist environment variables across a Claude session?

Accepted Answer

You can use the SessionStart hook to append export commands to the file path stored in $CLAUDE_ENV_FILE, which Claude will then load for subsequent operations.

Question 3

Can I run multiple hooks for the same event?

Accepted Answer

Yes, all matching hooks run in parallel. This allows you to combine independent checks, such as a fast deterministic bash script and a more nuanced prompt-based validation, for the same event.

Question 4

What are prompt-based hooks in Claude Code?

Accepted Answer

Prompt-based hooks use LLM reasoning to evaluate context-aware decisions, such as validating if a specific tool call is safe or if a task has been completed correctly, without requiring complex bash scripting.

Question 5

What is the difference between PreToolUse and Stop hooks?

Accepted Answer

PreToolUse triggers before a tool is executed to allow for validation or modification, while a Stop hook triggers when Claude intends to end the session, allowing you to verify that all requirements were met.

Claude Code Hook Development

Key Features

Use Cases

Claude Code Hook Development

Key Features

Use Cases