ui-color-palette-sketch
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill instructs the agent to use the
run_codetool inreferences/extract-source-colors.mdandreferences/extract-styles-colors.mdto execute JavaScript snippets. These scripts interact with the Sketch API to collect color data from layers and shared styles. The scripts are provided as static templates within the documentation, which mitigates the risk of arbitrary code generation. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external Sketch documents (such as layer names, color values, and style metadata), which introduces a potential attack surface.
- Ingestion points: Solid fill colors and shared style properties extracted via
run_codescripts from the current Sketch selection. - Boundary markers: The instructions do not define explicit delimiters or warnings to ignore potentially malicious instructions embedded in Sketch layer names or metadata.
- Capability inventory: The skill has the ability to execute code via
run_codeand perform document-level modifications using Sketch-specific tools (e.g., creating swatches, styles, and artboards). - Sanitization: There is no mention of sanitizing or validating extracted strings (like layer names) before they are used in downstream palette generation or UI rendering.
Audit Metadata