21st-ai
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
21stCLI tool (associated with the vendor package@21st-dev/cli) to execute various UI development tasks: 21st generate: Used to initiate UI sketching from a natural language prompt.21st generation: Used to list available variants of a generated project.21st iterate: Used to refine specific UI variants.21st take: Used to retrieve design specs or HTML code from the 21st platform.- [SAFE]: The documentation includes Russian language examples for natural language prompts (e.g., "сгенери UI на 21st", "поправь третий вариант", "забери код того наброска"). These are legitimate Cyrillic text examples for multi-lingual support and do not constitute homoglyph attacks or obfuscation.
- [SAFE]: Authentication instructions recommend standard secure practices, such as using
21st loginfor token management or using theTWENTYFIRST_TOKENenvironment variable. No hardcoded credentials or private keys were found in the skill content.
Audit Metadata