design-polish
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from project files that could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill reads project context from
.design-context.mdand prior critique snapshots from the.design-critique/directory to guide its polishing process (SKILL.md). - Boundary markers: There are no specific delimiters or instructions to ignore embedded commands within the ingested context files.
- Capability inventory: The skill explicitly "executes changes" (file system writes) to implement UI improvements, providing a mechanism for an injection to affect the codebase.
- Sanitization: The instructions do not define any validation or filtering logic for the content retrieved from project files.
Audit Metadata