unocss
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze project source files (e.g., .vue, .tsx, .html) and configuration files (uno.config.ts) to extract CSS utilities. This ingestion of potentially untrusted project content creates a surface for indirect instructions to influence agent behavior.
- Ingestion points: User project source code and configuration files as described in SKILL.md and references/core-extracting.md.
- Boundary markers: The skill recommends checking for configuration files to establish context in SKILL.md, but does not specify delimiters or "ignore instructions" directives for extracted content.
- Capability inventory: The agent generates CSS rules, modifies configuration objects, and applies shortcuts based on the extracted content.
- Sanitization: The skill does not prescribe specific sanitization or validation logic for data extracted from project files during the CSS generation process.
- [EXTERNAL_DOWNLOADS]: Documentation includes instructions for installing official packages from the NPM registry (unocss, @unocss/nuxt, @unocss/preset-icons) and references well-known development services such as Iconify and esm.sh. These are standard resources for the technology.
- [COMMAND_EXECUTION]: The skill provides command-line examples for setting up the environment using standard package managers (e.g., pnpm add -D ...).
Audit Metadata