Detect security vulnerabilities in Claude Code configurations with static analysis.
npx skills add https://github.com/ItsMattG/property-tracker --skill agent-shield
Please help me install this Agent Skill. Skill: agent-shield Source: https://github.com/ItsMattG/property-tracker/tree/main/.claude/skills/agent-shield Command: npx skills add https://github.com/ItsMattG/property-tracker --skill agent-shield
Scans CLAUDE configuration files for security vulnerabilities and misconfigurations, helping prevent credential exposure and policy violations.
Run a quick static scan of the .claude directory to generate a Markdown security report.
High-intent search queries and answers about installing and using this skill.
A deep security scan of Claude configurations uses a three-agent pipeline to simulate threats and produce remediation guidance, while a quick static scan analyzes hooks, skills, CLAUDE.md, and settings.json for vulnerabilities without deep simulation.
You need an ANTHROPIC_API_KEY specifically for deep security scans that utilize the three-agent Opus pipeline to simulate threats, whereas quick static analysis of configuration files does not require it.
The best way to prevent secrets leakage in Claude Code hooks and skills is to run automated pre-PR audits and post-change validation that detect misconfigurations and credential exposure across your configuration files.
You can generate a security report for CLAUDE.md and settings.json modifications by running an automated scan that outputs a Markdown report highlighting HIGH and CRITICAL findings with actionable remediation steps.
A Claude config audit detects MCP-related risks by scanning infrastructure modifications to prevent policy violations, insecure configurations, and credential exposure across hooks, skills, and settings files.
oabdelmaksoudScan Claude Code configurations for security vulnerabilities using AgentShield.
acta0724Analyze code changes and pull requests for security vulnerabilities.
hardw00tScan Infrastructure as Code configurations for misconfigurations and policy violations.
Obsidian-OwlScan changed files for leaked secrets and injection patterns using SAST tools.
