skills.rest
Science-Discovery
Science-DiscoveryOFFICIAL
Explore all skills by this author →
GitHub

skill-security-auditor

Audit AI Skill packages for security risks and output PASS/WARN/FAIL reports.

71|18|Updated Mar 6, 2026
Search Tags:#security#vulnerability#audit#scan#dependency#prompt-injection#ai-skills
One-click install
npx skills add https://github.com/Science-Discovery/Aether --skill skill-security-auditor-science-discovery
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-security-auditor
Source: https://github.com/Science-Discovery/Aether/tree/main/.opencode/skills/skill-security-auditor
Command: npx skills add https://github.com/Science-Discovery/Aether --skill skill-security-auditor-science-discovery

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Security auditing of AI agent skills before installation to prevent risks. This Skill scans SKILL.md prompts, scripts, dependencies, and filesystem usage to surface vulnerabilities and unsafe patterns.

Core Features & Use Cases

  • Pre-install security gates for AI skills and plugins, including prompt reviews and dependency checks.
  • Detect dangerous patterns in Python scripts (e.g., os.system, eval, subprocess with shell) and risky dependencies.
  • Scan SKILL.md for prompt injection indicators and validate references for safety.
  • Produce actionable PASS/WARN/FAIL verdicts with remediation guidance for each finding.

Quick Start

Run the auditor on a local skill directory to generate a security report.

Frequently Asked Questions about skill-security-auditor

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

FAQPage Schema
How do I scan an AI agent skill for prompt injection vulnerabilities?▼

To scan for prompt injection vulnerabilities, you audit the SKILL.md file and associated scripts to detect unsafe patterns and validate references. This process surfaces malicious prompts and outputs a structured PASS/WARN/FAIL security verdict with remediation guidance.

How do I check local skill directories and git repositories for security risks before installation?▼

Checking local directories and git repositories for security risks requires scanning SKILL.md prompts, scripts, dependencies, and filesystem usage. This pre-install audit identifies vulnerabilities and generates a report with actionable remediation guidance.

What dangerous Python patterns should I look for in AI skill scripts?▼

Dangerous Python patterns in AI skill scripts include os.system, eval, and subprocess calls with shell enabled. Scanning for these risky dependencies and unsafe commands helps prevent code execution vulnerabilities before installation.

Does the security audit evaluate filesystem usage by AI plugins?▼

Yes, the security audit evaluates filesystem usage by AI plugins to identify unsafe access patterns. It scans how the skill interacts with the local filesystem to surface potential risks and assigns a PASS, WARN, or FAIL verdict.

What is the best way to generate a PASS or FAIL security report for an AI skill package?▼

The best way to generate a security report is to run an automated audit on the AI skill package. This produces a structured verdict with PASS, WARN, or FAIL status and provides specific remediation guidance for each detected vulnerability.

Related Skills

Explore More Education & Research Skills→
heliosarchitectheliosarchitect
1

skill-name

Generate an OpenClaw skill scaffold with SKILL.md frontmatter and optional directories.

Community
Basic
theandyalvarez7-rubytheandyalvarez7-ruby

skill-security-auditor

Audit AI agent skills for security risks and generate JSON reports.

Community
Advanced
thomast1906thomast1906
19

terraform-security-scan

Automate security scanning of Terraform configurations for Azure infrastructure.

Community
Advanced
akaszubskiakaszubski
32

api-integration-patterns

Integrate external APIs and CLI tools with secure subprocess calls and retry logic.

Community
Advanced