audit
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external content from the codebase being audited and project-specific configuration files such as
AGENTS.mdandCLAUDE.md. - Ingestion points: Project source code, diffs, and local documentation files.
- Boundary markers: The instructions do not explicitly mandate the use of delimiters (e.g., XML tags or triple backticks) to separate untrusted code content from the agent's internal reasoning or instructions.
- Capability inventory: Across its scripts, the skill performs local file read operations and writes output to a local directory (
.audit/), including generating a JavaScript data file (report.js). - Sanitization: The accompanying
report.htmltemplate primarily usestextContentfor DOM manipulation, which provides a layer of protection against Cross-Site Scripting (XSS) when rendering audit findings. - [EXTERNAL_DOWNLOADS]: The reporting template fetches font styles and assets from Google's official public font repositories.
Audit Metadata