How do I install the Hook Block Commands skill?

Run npx skillmds add yanacuti1121/hook-block-commands in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Hook Block Commands safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Hook Block Commands?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Hook Block Commands free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Hook Block Commands?

yanacuti1121 (@yanacuti1121) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Hook Block Commands by yanacuti1121

Name: Hook Block Commands
Rating: 4.2 (5 reviews)
Author: yanacuti1121

Hook Block Commands

Pattern guide for writing PreToolUse hooks that block dangerous shell commands. Use when: building or auditing a guard-destructive hook, adding new blocked patterns, reviewing what the current hook covers. Covers 3 safety levels (critical/high/strict) and 58+ regex patterns. Inspired by: karanb192/claude-code-hooks block-dangerous-commands pattern (MIT).

yanacuti1121 4e012dc 2.3 KB Updated Sep 9, 2026 2 repo stars

File contents

Hook Block Commands Pattern

When to trigger

"what does guard-destructive block?"
"add pattern to hook", "block this command"
"review hook coverage", "hook pattern"
Auditing core/hooks/guard-destructive.sh

Safety levels

Level	Scope
`critical`	Catastrophic only: rm -rf ~, dd to disk, fork bombs
`high` (default)	+ force push main, git reset --hard, git clean -f, chmod 777, curl\|sh, secret file ops
`strict`	+ any force push, sudo rm, docker prune, crontab -r

Core pattern categories (58+ patterns)

Filesystem destruction

rm -rf /, rm -rf ~, rm -rf $HOME
mkfs.*, dd if=.* of=/dev/
shred, wipe

Git destructive

git push.*--force.*main, git push.*-f.*main
git reset --hard, git clean -fdx, git filter-branch
git reflog expire

System/process

:(){:|:&};: (fork bomb)
chmod -R 777, chmod 777 /
sudo rm, sudo chmod

Secret exposure

cat ~/.aws/credentials, cat .env
echo $.*TOKEN, env | grep.*KEY
Piping secrets to curl/netcat

Exfiltration

curl.*upload, scp.*, rsync.*remote
nc.*<IP>, netcat.*

Hook exit codes

exit 0  → allow (no output)
exit 2  → block (JSON reason on stdout)

Blocking response format

{
  "decision": "block",
  "reason": "🚫 Blocked: rm -rf on home directory. Safety level: high."
}

Adding new patterns to guard-destructive.sh

Identify the pattern category (filesystem / git / system / secret / exfiltration)
Write a regex that matches the dangerous form but not safe variants
Add to the correct safety level block
Add a test case to core/tests/hooks/run-hook-tests.sh
Update hook # Last Reviewed: date

Reference

YAMTAM hook: core/hooks/guard-destructive.sh Tests: core/tests/hooks/run-hook-tests.sh

---name: hook-block-commandsdescription: Pattern guide for writing PreToolUse hooks that block dangerous shell commands. Use when: building or auditing a guard-destructive hook, adding new blocked patterns, reviewing what the current hook covers. Covers 3 safety levels (critical/high/strict) and 58+ regex patterns. Inspired by: karanb192/claude-code-hooks block-dangerous-commands pattern (MIT).--- # Hook Block Commands Pattern ## When to trigger - "what does guard-destructive block?" - "add pattern to hook", "block this command" - "review hook coverage", "hook pattern" - Auditing `core/hooks/guard-destructive.sh` ## Safety levels | Level | Scope | |---|---| | `critical` | Catastrophic only: rm -rf ~, dd to disk, fork bombs | | `high` (default) | + force push main, git reset --hard, git clean -f, chmod 777, curl\|sh, secret file ops | | `strict` | + any force push, sudo rm, docker prune, crontab -r | ## Core pattern categories (58+ patterns) ### Filesystem destruction - `rm -rf /`, `rm -rf ~`, `rm -rf $HOME` - `mkfs.*`, `dd if=.* of=/dev/` - `shred`, `wipe` ### Git destructive - `git push.*--force.*main`, `git push.*-f.*main` - `git reset --hard`, `git clean -fdx`, `git filter-branch` - `git reflog expire` ### System/process - `:(){:|:&};:` (fork bomb) - `chmod -R 777`, `chmod 777 /` - `sudo rm`, `sudo chmod` ### Secret exposure - `cat ~/.aws/credentials`, `cat .env` - `echo $.*TOKEN`, `env | grep.*KEY` - Piping secrets to curl/netcat ### Exfiltration - `curl.*upload`, `scp.*`, `rsync.*remote` - `nc.*<IP>`, `netcat.*` ## Hook exit codes ```bash exit 0 → allow (no output) exit 2 → block (JSON reason on stdout) ``` ## Blocking response format ```json { "decision": "block", "reason": "🚫 Blocked: rm -rf on home directory. Safety level: high." } ``` ## Adding new patterns to guard-destructive.sh 1. Identify the pattern category (filesystem / git / system / secret / exfiltration) 2. Write a regex that matches the dangerous form but not safe variants 3. Add to the correct safety level block 4. Add a test case to `core/tests/hooks/run-hook-tests.sh` 5. Update hook `# Last Reviewed:` date ## Reference YAMTAM hook: `core/hooks/guard-destructive.sh` Tests: `core/tests/hooks/run-hook-tests.sh`

Frequently asked questions