touch-ui
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to executepython3 -m touch_agent.uiux. This vendor-specific module is used to query a knowledge base using parameters like{platform},{topic}, and{framework}. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection via shell command interpolation:
- Ingestion points: User-provided inputs for platform, topic, and framework are directly interpolated into the search commands in
SKILL.md. - Boundary markers: There are no markers or instructions defined to delimit user input or prevent command injection.
- Capability inventory: The skill has access to the
Bashtool for executing arbitrary shell commands. - Sanitization: There is no explicit sanitization or validation logic described for the processed data.
Audit Metadata