audit-ux-journeys
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to ingest and process untrusted data from local source code and external websites.
- Ingestion points: Local file content parsed via
rgandfdcommands, and remote website content processed during Phase 2 browser walkthroughs (SKILL.md, checklist.md). - Boundary markers: The skill does not define explicit delimiters or instructions to ignore potential malicious prompts embedded within the files or websites being audited.
- Capability inventory: The agent uses shell commands (
fd,rg,sed) and a headed browser for navigation and screenshot capture. - Sanitization: No explicit sanitization or filtering is performed on the content extracted from the target application.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (
fd,rg,sed) to perform route inventory, search for navigation patterns, and identify analytics configurations (Phase 0, Phase 3). These commands are standard for technical auditing and are restricted to searching the local project directory.
Audit Metadata