21st-ui
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches React component source code, installation metadata, and brand logos from the official 21st.dev API (https://21st.dev/api/mcp).
- [COMMAND_EXECUTION]: Instructions guide the agent to modify the local project by writing component files and executing installation commands such as
shadcn-style registry add. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external API responses which could theoretically contain instructions intended to influence the agent. \n- Ingestion points: Remote data enters the context through tool outputs from
search,get_component, andgenerate. \n- Boundary markers: There are no explicit delimiters or safety instructions provided to the agent regarding the fetched remote content. \n- Capability inventory: The agent is authorized to write files to the project directory and execute shell commands for component setup. \n- Sanitization: The instructions do not mandate validation or sanitization of the content returned by the API before it is applied.
Audit Metadata