stitch-ui-vue-layui-components
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the shell script
scripts/fetch-stitch.shusingbash. This script usescurlto download HTML content from a URL provided by the external Stitch service. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted HTML and JSON metadata from Stitch design screens (Ingestion point:
get_screenoutput in SKILL.md). The workflow lacks explicit boundary markers or sanitization instructions to prevent adversarial content within the designs from influencing the agent. The agent possesses capabilities to write files and execute local scripts (Capability:bash scripts/fetch-stitch.sh,npm install). - [EXTERNAL_DOWNLOADS]: The skill facilitates the download of design assets from Google Cloud Storage URLs provided by the Stitch platform.
- [REMOTE_CODE_EXECUTION]: The skill directs the agent to run
npm installfor project setup, which involves downloading and executing third-party code from the npm registry.
Audit Metadata