responsive-ui-qa
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it instructs the agent to process and act upon untrusted code and documentation from a repository.
- Ingestion points: The agent reads repository instructions and inspects rendered frontend components to diagnose defects (SKILL.md, Workflow steps 1 and 3).
- Boundary markers: No instructions are provided to the agent to treat external content as untrusted or to use specific delimiters for data/instruction separation.
- Capability inventory: The skill requires starting the application via shell commands and modifying files to implement fixes (SKILL.md, Workflow steps 1 and 6).
- Sanitization: There are no specified procedures for validating or sanitizing content ingested from the external repository.
- [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands provided by the project being tested to start the application.
- Evidence: Workflow Step 1 in SKILL.md states: 'Start the app with its normal command.'
Audit Metadata