claude-code-hook-development

This skill should be used when the user asks to "create a hook", "add a hook", "write a hook", or mentions Claude Code hooks. Also suggest this skill when the user asks to "automatically do X" or "run X before/after Y" as these are good candidates for hooks.

Invalid

This skill can't be scored yet

Validation errors are blocking scoring. Review and fix them to unlock Quality, Impact and Security scores. See what needs fixing →

Evaluation results

94%

13%

Protecting a Repository from Destructive Shell Commands

PreToolUse command blocking hook

Criteria

Without context

With context

Settings file location

100%

Script file location

100%

Script is executable

Bash shebang

100%

Pipefail option

100%

Stdin JSON parsing

70%

100%

Exit 2 for blocking

100%

Stderr for block message

100%

Case-sensitive Bash matcher

100%

PreToolUse event

100%

Hook type is command

100%

Exit 0 for allowed commands

100%

Non-blocking error exit

100%

92%

21%

Automating Code Quality Feedback After Every File Edit

PostToolUse quality enforcement hook

Criteria

Without context

With context

PostToolUse event used

100%

Edit matcher

100%

Settings file location

100%

Script file location

100%

Script is executable

Bash shebang

100%

Pipefail option

100%

File path extraction

80%

100%

No exit 2 blocking

100%

Feedback to stdout

100%

70%

File type filtering

100%

Hook type is command

100%

Non-zero exit does not block

100%

95%

Diagnosing and Fixing a Broken Pre-Push Hook Setup

Hook lifecycle and activation workflow

Criteria

Without context

With context

Restart or /hooks required

100%

Snapshot-at-startup explanation

100%

JSON validation command

100%

Debug logging technique

100%

Case-sensitive matcher check

100%

Early exit for non-push commands

100%

Exit 2 blocking with stderr

100%

Settings file location

100%

Script file location

100%

Script executable

40%

Timeout configuration

100%

Stdin JSON parsing with jq

100%

disableAllHooks check

100%

Repository: dwmkerr/claude-toolkit
Commit: 92f335c

Evaluated: 19 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Protecting a Repository from Destructive Shell Commands Automating Code Quality Feedback After Every File Edit Diagnosing and Fixing a Broken Pre-Push Hook Setup

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.