stitch-uview-plus-components

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local bash script scripts/fetch-stitch.sh to handle the download of design assets, which is a standard procedure for its intended conversion task.
  • [EXTERNAL_DOWNLOADS]: The skill fetches HTML design source from stitch.withgoogle.com, a well-known service. These downloads are performed using curl to retrieve the visual source of truth required for the conversion.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it processes external design data.
  • Ingestion points: External HTML content is downloaded to temp/source.html for analysis and conversion.
  • Boundary markers: There are no specific instructions or delimiters provided to prevent the agent from potentially following instructions embedded within the design HTML.
  • Capability inventory: The skill uses Bash and file system tools (Read, Write) to manage the conversion workflow.
  • Sanitization: No validation or sanitization of the downloaded HTML is defined before the agent parses the content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 04:04 PM
Security Audit — agent-trust-hub — stitch-uview-plus-components