dioxus-stitch

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to use the Bash tool for fetching raw HTML/Tailwind and to execute a specific Python script (skills/dioxus-ui-ux/scripts/search.py) for component lookups.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from external sources.
  • Ingestion points: Data is retrieved via [prefix]:get_screen and screenshot.downloadUrl from the Stitch MCP tool.
  • Boundary markers: There are no specified delimiters or instructions to treat the ingested design content as untrusted data.
  • Capability inventory: The agent has access to the Bash tool for shell command execution and filesystem access for creating Rust source files in src/components/.
  • Sanitization: The instructions do not define any validation or sanitization steps for the data retrieved from the Stitch design before it is used to generate code or passed to tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 10:20 AM