skills.rest
tunnckoCore
tunnckoCoreCOMMUNITY
Explore all skills by this author →
Website𝕏@wgw_ethGitHub

agent-security

Scan git history and files for exposed secrets and enforce security rules.

Updated Feb 28, 2026
Search Tags:#security#audit#secrets#prompt injection#guardrails#hardening
One-click install
npx skills add https://github.com/tunnckoCore/agent-skills --skill agent-security-tunnckocore
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-security
Source: https://github.com/tunnckoCore/agent-skills/tree/main/skills/agent-security
Command: npx skills add https://github.com/tunnckoCore/agent-skills --skill agent-security-tunnckocore

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node, git, grep, find, chmod, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides essential security guardrails, self-audit tools, and secret scanning to protect AI agents from compromise, credential leaks, and manipulation.

Core Features & Use Cases

  • Non-negotiable Rules: Enforces strict security policies that cannot be overridden.
  • Self-Auditing: Runs scripts to check file permissions, git history for secrets, and network exposure.
  • Secret Scanning: Detects accidentally committed API keys, private keys, and tokens.
  • Use Case: An agent is about to execute a script that might leak its API keys. This skill intercepts the execution, flags the potential leak, and prevents the action, alerting the human operator.

Quick Start

Run the security audit script to check your agent's current security posture.

Frequently Asked Questions about agent-security

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent prompt injection attacks and credential exfiltration in AI agents?▼

Secret scanning for AI agents detects accidentally committed API keys, tokens, and private keys by analyzing your git history and current workspace files, flagging potential credential leaks before execution.

Do I need node.js and git to run automated security audits on my agent?▼

Yes, running automated security audits requires node.js, git, and grep for script execution and analysis. These dependencies are necessary to check file permissions, network exposure, and scan git history for secrets.

How do I enforce strict security policies that an AI agent cannot override?▼

You can enforce strict security policies by applying non-negotiable rules that the agent cannot bypass. These guardrails intercept script executions that might leak API keys or perform unauthorized transactions, alerting the human operator.

What is the best way to scan git history for committed API keys and tokens?▼

The best way to scan git history for committed API keys is using automated secret scanning tools that detect private keys and tokens, preventing accidental credential leaks by alerting operators before exposure.

Can I intercept an AI agent script execution before it leaks sensitive data?▼

Yes, you can intercept script execution before it leaks sensitive data by implementing security guardrails that flag potential API key leaks, prevent the action, and alert the human operator immediately.

Related Skills

Explore More Software Engineering Skills→
hve4638hve4638

frame-setup

Install the @ast-grep/napi runtime dependency globally via npm.

Community
Intermediate
guccangguccang
3

webhook-subscriptions

Create webhook subscriptions that trigger agent runs from external events.

Community
Intermediate
mitul-bhatiamitul-bhatia

perl-security

Harden Perl applications with taint mode, input validation, and parameterized DBI queries.

Community
Advanced
abzhawabzhaw

security-scanning

Automate dependency audits, SAST, and credential leak detection across Julia's codebase.

Community
Advanced