storyboard
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
prettier --writeto format the generated storyboard HTML files. This is a standard and expected use of a well-known development utility for maintaining code style. - [EXTERNAL_DOWNLOADS]: No remote script downloads or external package installations from untrusted sources were detected. The skill primarily interacts with the local file system and internal agent tools.
- [DATA_EXFILTRATION]: The skill reads project documentation and design mockups to populate the storyboard. All output is directed to local HTML files within the repository structure (e.g.,
design-mocks/), with no evidence of network transmission or exfiltration to external domains. - [PROMPT_INJECTION]: The instructions do not contain patterns aimed at overriding agent constraints or safety guidelines. It follows a structured workflow for gathering inputs and producing a specific artifact.
- [SAFE]: The functionality of the skill aligns with its documented purpose. It integrates with existing project workflows by updating manifests and referencing existing design tokens, following best practices for development-oriented agent skills.
Audit Metadata