21st-ui-build
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the 21st command-line interface to perform several actions including initializing design context (21st init), searching for components (21st search), generating UI variants (21st generate), and reviewing changes (21st review). These operations are consistent with the skill's stated purpose of UI automation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could potentially contain malicious instructions.
- Ingestion points: Reads project-specific design information from .21st/design.json and .21st/DESIGN.md, and incorporates output from 21st search commands.
- Boundary markers: The instructions lack explicit boundary markers or delimiters to differentiate between design data and behavioral instructions.
- Capability inventory: The agent has the capability to execute shell commands via the 21st CLI and modify project source files to implement UI components.
- Sanitization: There is no evidence of sanitization or validation performed on the ingested design metadata or external search results before they are processed by the agent.
Audit Metadata