brooks-health

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted project data, including source code files, entry points, and version control diffs, to generate a health dashboard. This creates a standard surface for indirect prompt injection where malicious content in the analyzed codebase could attempt to manipulate the agent's output.
  • Ingestion points: Reads project structure, import statements, test files, and git diffs during the scanning process defined in health-guide.md.
  • Boundary markers: The process lacks explicit delimiters to separate the codebase content from the analysis instructions, relying on the agent's internal reasoning to differentiate.
  • Capability inventory: The skill is limited to reading local project files and generating text/graphical reports. It does not possess network transmission or file-write capabilities in its described process.
  • Sanitization: The instructions do not specify sanitization or escaping of project content before it is summarized in the dashboard findings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 07:28 PM
Security Audit — agent-trust-hub — brooks-health