ux-journey-architect
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The
SKILL.mdfile contains instructions to execute external Python scripts for skill validation (lint_production_skill.pyandlint_skill_cluster.py). These scripts are referenced via environment variables$CODEX_HOMEand$HOMEand are not provided within the skill package, meaning their exact behavior is unverifiable if executed by the agent. - [PROMPT_INJECTION]: The skill's core function involves processing site structures and audience data, which introduces a surface for indirect prompt injection.
- Ingestion points: Site goals, audience segments, and URL maps provided in user prompts or derived from external site descriptions.
- Boundary markers: The workflow lacks explicit delimiters or instructions to ignore potential commands embedded within the data being analyzed.
- Capability inventory: The skill is primarily instructional for generating Markdown journey maps; it does not invoke dangerous tools or perform network operations during the journey mapping process.
- Sanitization: No sanitization or filtering of external input is defined in the workflow.
Audit Metadata