Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Andreas Asprou avatar

Safety Net

  • 7 repo stars
  • Updated June 21, 2026
  • andreasasprou/agent-skills

Install a guardrail that blocks destructive shell commands before your coding agent executes them during any repo or infra task.

About

Safety Net is an agent plugin skill that stops destructive shell commands before they run, giving developers a default guardrail when Claude Code, Cursor, or similar agents touch git, filesystem, or cloud CLIs. Instead of trusting the model to refuse risky suggestions, the plugin enforces blocks on patterns such as recursive deletes, force pushes, and AWS resource deletions called out in the skill description. It fits developers who automate infra and cleanup via agents but cannot afford one mistaken rm -rf or forced git history rewrite. Complexity is low: install the plugin and let pre-execution hooks filter commands while you still review Security Audits on the catalog page. It does not replace backups, branch protection, or least-privilege IAM—it narrows the blast radius of agent-initiated terminal mistakes across every phase where a shell is used.

  • Intercepts shell commands before execution and blocks destructive patterns
  • Covers rm -rf, git push --force, AWS deletions, and related high-risk operations
  • Plugin sourced from ./plugins/safety-net in the agent-skills repo
  • Lightweight policy layer—no substitute for human review of still-risky commands
  • Works as a safety net across build, ship, and operate terminal workflows

Safety Net by the numbers

  • Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add andreasasprou/agent-skills
/plugin install safety-net@andreas-agent-skills

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
repo stars7
Last updatedJune 21, 2026
Repositoryandreasasprou/agent-skills

What it does

Install a guardrail that blocks destructive shell commands before your coding agent executes them during any repo or infra task.

Files

SKILL.mdMarkdownGitHub ↗

Block destructive shell commands before execution. Prevents rm -rf, git push --force, AWS deletions, and more.

safety-net

Source: ./plugins/safety-net

{ "name": "safety-net", "source": "./plugins/safety-net", "description": "Block destructive shell commands before execution. Prevents rm -rf, git push --force, AWS deletions, and more." }

Related skills

Securityappsecsecrets

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.