shadcn-ui
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The instructions use high-priority markers ("CRITICAL", "MUST", "Do not proceed without") to force the agent to disregard its internal training data in favor of external content. This is a form of system prompt override, although it is directed towards a legitimate technical source.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data via the
WebFetchtool from a remote source (https://ui.shadcn.com/llms.txt), which could influence the agent's output if the remote file contains malicious instructions. - Ingestion points:
SKILL.mdexplicitly directs the agent to callWebFetchon a remote URL. - Boundary markers: Absent. The skill does not provide delimiters or instructions to the agent to isolate the fetched documentation from its own control logic.
- Capability inventory: The agent uses the fetched information to generate React components and suggest CLI commands (
shadcn add). - Sanitization: Absent. The skill does not define any validation for the content returned by the remote fetch.
Audit Metadata