namethatui
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill includes internal Python scripts (test_skill.py, validate.py) that use subprocess.run to execute local helper scripts (prepare_research.py) for automated testing and package validation. These executions are restricted to static paths within the skill directory and do not process unsanitized user input.\n- [EXTERNAL_DOWNLOADS]: The skill is designed to research documentation from established sources such as W3C and MDN. It incorporates a specialized URL guard (prepare_research.py) that prevents navigation to a specific blocked domain (namethatui.com) as a matter of policy and grounding.\n- [PROMPT_INJECTION]: The skill analyzes external web content and DOM structures, presenting an indirect prompt injection surface. Ingestion points include live pages and accessibility trees as defined in SKILL.md. Boundary markers are present in the form of instructions to separate observed facts from inferred behavior and using separate ledgers for evidence. Capabilities include internal shell execution for testing and browser control. Sanitization is performed via destination URL filtering.
Audit Metadata