style-library
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Node.js scripts (
new-style.mjsandbuild-registry.mjs) located in the project'sscripts/directory to automate the creation of new styles and the updating of the style registry. This is consistent with the skill's stated purpose of managing a library. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from various project files which represents a potential attack surface.
- Ingestion points: The agent reads external content from
style-library/registry.json,DESIGN.md,tokens.css, andstyle-templates/README.md. - Boundary markers: No explicit boundary markers or "ignore embedded instructions" directives are present for these ingestion points.
- Capability inventory: The skill has the capability to execute Node.js scripts and perform file system operations (copying assets to
compositions/). - Sanitization: No explicit sanitization or validation of the content within the ingested JSON, CSS, or Markdown files is mentioned.
Audit Metadata