draft-ux
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external project files to drive the generation of UI prototypes and documentation. * Ingestion points: Untrusted data is read from 'requirements.md' and 'ui-brief.md' (SKILL.md Step 1, 5). * Boundary markers: Absent; the skill lacks specific delimiters or instructions to ignore potential commands embedded in the requirement files. * Capability inventory: The skill possesses the ability to write to 'ui-brief.md', delete files on the local system, and execute 'git status' (SKILL.md Step 5, 6). * Sanitization: No sanitization, escaping, or validation is performed on the ingested requirements before they influence agent logic.
- [COMMAND_EXECUTION]: The skill requires the agent to execute 'git status' and perform file system deletions to ensure prototype artifacts are removed after a design is locked (SKILL.md Step 6).
Audit Metadata