enhance-prompt
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the ingestion of external project files.\n
- Ingestion points: The skill reads the
DESIGN.mdfile inSKILL.mdto extract design system context and tokens.\n - Boundary markers: The extracted content is placed under a structured header, but the instructions do not explicitly command the agent to ignore or isolate instructions that may be embedded within the project file.\n
- Capability inventory: The skill has access to
ReadandWritetools to manage design documentation and generate output prompts.\n - Sanitization: No specific sanitization or schema validation is performed on the content read from the
DESIGN.mdfile.\n- [EXTERNAL_DOWNLOADS]: The skill references up-to-date prompting guidelines from Google's official documentation at stitch.withgoogle.com, which is a well-known and trusted service.
Audit Metadata