stitch-uviewpro-components

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a Bash script (scripts/fetch-stitch.sh) to perform downloads. This is a common pattern for reliable content retrieval.
  • [EXTERNAL_DOWNLOADS]: The skill fetches design specifications and HTML content from Stitch (a Google service) via the get_screen tool and a helper script.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external HTML content from design files.
  • Ingestion points: External HTML content is fetched from URLs provided by the Stitch design tool (SKILL.md, scripts/fetch-stitch.sh).
  • Boundary markers: No specific delimiters or instructions to disregard embedded commands in the source HTML are present in the provided files.
  • Capability inventory: The agent has permissions for Bash, Write, Read, and web_fetch tools.
  • Sanitization: The fetched HTML is processed directly to generate code without a visible sanitization or validation step.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 10:44 AM