stitch-code-to-design
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references an upstream code snapshot from a Google-affiliated GitHub repository (google-labs-code/stitch-skills) to define its implementation logic. This is a reference to a well-known and trusted service.
- [COMMAND_EXECUTION]: The workflow requires executing standard development commands (e.g., npm run dev, ng serve) and installing related skills using npx skills add full-stack-skills/stitch-skills. These operations involve vendor-owned resources and are necessary for the skill's primary function.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user source code (HTML, components) as ingestion points (SKILL.md). It manages this risk through boundary markers like defined route lists, a clear inventory of capabilities including file-writing and network upload (references/workflow.md), and mandatory sanitization steps (SKILL.md, references/anti-patterns.md) that replace sensitive data with demo values and remove credentials before processing.
- [CREDENTIALS_UNSAFE]: The skill demonstrates safe secret management by explicitly instructing the agent to use environment variables for the STITCH_API_KEY and forbidding the inclusion of keys or tokens in logs or reports.
Audit Metadata