OpenSkillIndex← back to search
claude-code

Llmops-demo-ts security-audit

Perform a security audit of the codebase. Checks for OWASP Top 10, AI-specific vulnerabilities, dependency issues, and configuration problems.

install
source · Clone the upstream repo
git clone https://github.com/yu-iskw/llmops-demo-ts
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/yu-iskw/llmops-demo-ts "$T" && mkdir -p ~/.claude/skills && cp -r "$T/.claude/skills/security-audit" ~/.claude/skills/yu-iskw-llmops-demo-ts-security-audit && rm -rf "$T"
manifest: .claude/skills/security-audit/SKILL.md
source content

Security Audit

Perform a security audit with the following scope:

$ARGUMENTS

Audit Methodology

1. Dependency Security

pnpm audit

Review all known vulnerabilities in dependencies.

2. Source Code Analysis

Scan for common vulnerability patterns:

  • Hardcoded secrets (API keys, passwords, tokens)
  • Command injection via string interpolation in Bash/exec calls
  • XSS vectors in Vue.js templates (v-html usage)
  • Prompt injection in AI agent inputs
  • Insecure deserialization
  • Information disclosure in error messages

3. AI Agent Security

Review the secure_agent pattern and verify:

  • Input sanitization is applied before LLM processing
  • Output sanitization prevents data leakage
  • Tool calls are validated and scoped
  • Prompt injection defenses are in place

4. Configuration Security

  • No secrets in version control
  • Proper .gitignore coverage
  • CORS configuration
  • Environment variable handling

Output

Produce a security report with findings classified by severity:

  • 🔴 Critical / 🟠 High / 🟡 Medium / 🔵 Low

Each finding includes: location, vulnerability, impact, and remediation steps.