stitch-ui-design
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied design tasks and context, which creates an attack surface for indirect prompt injection if the input contains malicious instructions.
- Ingestion points: User-provided task descriptions and context, as highlighted in the prompt templates within
SKILL.mdandreferences/prompt-examples.md. - Boundary markers: The instructions lack explicit delimiters or warnings for the agent to ignore instructions that might be embedded within the user's design requirements.
- Capability inventory: The skill is associated with high-privilege tools in
metadata.json, includingclaude-code,gemini-cli, andcodex-cli, which have access to the file system and shell execution. - Sanitization: There is no evidence of input validation or sanitization to prevent untrusted content from influencing the agent's logic.
Audit Metadata