ui-color-palette-sketch

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the run_code tool to execute local JavaScript snippets in references/extract-source-colors.md and references/extract-styles-colors.md. These scripts utilize the sketch module to programmatically access the current document's selection and shared styles to extract color data. This is a standard implementation for Sketch automation.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads color hex values and style names from the user's active Sketch document. This data is used to populate local ColorConfiguration and PaletteData objects within the agent's context and is not transmitted to external or untrusted domains.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from Sketch documents (layer names and style names) which could theoretically be manipulated by an attacker to include malicious instructions.
  • Ingestion points: Data is extracted via the run_code scripts in references/extract-source-colors.md and references/extract-styles-colors.md.
  • Boundary markers: The workflow includes a human-in-the-loop step (Step 2 in both extraction files) where the agent presents the extracted colors and names to the user for role assignment, acting as a manual verification boundary.
  • Capability inventory: The skill has the capability to write to the Sketch document and execute local code via run_code.
  • Sanitization: Extracted data is parsed as JSON and mapped into structured configuration objects with specific field validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 04:23 PM
Security Audit — agent-trust-hub — ui-color-palette-sketch