refactor(hooks-plugin): extract hook skill inline commands to standalone scripts

[laurigates]

Problem

Three skills in hooks-plugin use 4-5 shell utility Bash patterns (test, jq, cat, mkdir, chmod) in their allowed-tools frontmatter. This forces Claude to generate inline bash commands that each require individual user approval, since complex/compound commands can't be allowlisted via Bash(command *) patterns.

Affected skills

Skill	Shell utility patterns	Count
hooks-permission-request-hook	test, jq, cat, mkdir, chmod	5
hooks-session-start-hook	test, jq, cat, mkdir, chmod	5
hooks-session-end-issue-hook	test, jq, mkdir, chmod	4

Expected behavior

Consolidate shell utility operations into standalone scripts in each skill's scripts/ directory and use Bash(bash *) permission pattern. Since all three skills share similar patterns (hook creation/validation), consider shared scripts at the plugin level.

Reference

• See health-plugin/skills/health-check/ for the correct pattern
• Rule: .claude/rules/agentic-permissions.md → "Scripts for Compound Operations"
• Detected by: scripts/plugin-compliance-check.sh Check 6 (check_bash_patterns)

[laurigates]

Verified against current main at 179521b: all three hooks-plugin skills (hooks-permission-request-hook, hooks-session-start-hook, hooks-session-end-issue-hook) now declare allowed-tools: Read, Write, Edit, Glob, Grep, Bash, TodoWrite (unrestricted Bash, no shell-utility patterns) and contain no inline utility calls in their bodies — they are hook-template generators, not runners. The issue description predates a prior fix. Closing as stale.