design-qa
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several local Python and Node.js scripts to perform linting, token validation, and rendering analysis.
- Evidence: Execution of
validate_tokens.py,validate_contrast.py, andlint_hardcodes.pyviapython3. - Evidence: Execution of
measure_render.mjsvianodewhich utilizes headless Chromium for render-based contrast checks. - [INDIRECT_PROMPT_INJECTION]: The skill processes external source code and HTML components which could potentially contain malicious instructions intended to influence the agent during the auditing process.
- Ingestion points: Source code directories passed to
lint_hardcodes.pyand HTML files passed tomeasure_render.mjs(SKILL.md). - Boundary markers: None identified; the skill instructions do not specify delimiters or warnings for the agent when processing these files.
- Capability inventory: The skill has the ability to execute local scripts (Python/Node.js) and render content in a headless browser (SKILL.md).
- Sanitization: No explicit sanitization or validation of the ingested code/HTML is described before it is passed to the analysis scripts or rendering engine.
Audit Metadata