add-health-check
Warn
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: MEDIUMDATA_EXFILTRATIONPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DATA_EXFILTRATION]: The skill is designed to evaluate sensitive project data such as environment variables and secrets. Evidence: SKILL.md specifies that the 'infrastructure-check' analyzes 'env vars' and the 'security-analyzer' evaluates 'secrets'. Context: While this involves accessing sensitive information, it is a core function of the audit tool, and results are stored locally in the workspace.
- [PROMPT_INJECTION]: The skill ingests untrusted project content which could contain malicious instructions designed to influence the audit results. Ingestion points: Reads CLAUDE.md, infrastructure configurations, and project code. Boundary markers: No specific delimiters are used to isolate analyzed content. Capability inventory: The agent creates directories and writes multiple markdown reports. Sanitization: No data sanitization is specified.
- [COMMAND_EXECUTION]: The skill instructions involve file system operations to organize report outputs. Evidence: The 'Usage & Output' section describes creating dated directory structures (e.g., docs/health-checks/YYYY-MM-DD/) and generating several markdown analysis files.
Audit Metadata