stitch-vue-vant-components
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses a bash script
scripts/fetch-stitch.shto download design assets from URLs provided by the Stitch MCP tool. These downloads are required for the skill to retrieve the source HTML needed for conversion to Vue components.- [COMMAND_EXECUTION]: Uses theBashtool to execute a local script for fetching design data and performs standard project initialization and development commands such asnpm installandnpm run dev.- [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to ingest and process external HTML content to generate frontend code, which creates a potential vector for malicious instructions embedded in the source design data. - Ingestion points: External HTML data downloaded to
temp/source.htmland design metadata from Stitch. - Boundary markers: The instructions do not define clear delimiters or specific instructions for the agent to ignore embedded commands within the downloaded HTML.
- Capability inventory: The agent has permissions for file system writes, Bash command execution, and network access.
- Sanitization: There is no specified logic for sanitizing or validating the contents of the design HTML before it is processed by the AI.
Audit Metadata