stitch-vue-layui-components
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses a bash script to fetch design assets (HTML and metadata) from external URLs, specifically Google Cloud Storage hosted content associated with the Stitch platform.
- [COMMAND_EXECUTION]: The agent is instructed to execute a local bash script,
scripts/fetch-stitch.sh, which usescurlto perform network operations. While intended for asset retrieval, this provides a mechanism for local command execution. - [INDIRECT_PROMPT_INJECTION]: The skill ingests design data from external HTML files to generate code, creating an attack surface for indirect prompt injection.
- Ingestion points: The file
temp/source.htmlis used to store downloaded design content from external URLs. - Boundary markers: The instructions provide architectural constraints and a conversion checklist, but do not explicitly direct the agent to ignore instructions embedded within the source design HTML.
- Capability inventory: The skill has the ability to execute shell scripts (
scripts/fetch-stitch.sh) and write multiple files to the local file system (Vue components, mock data). - Sanitization: No explicit sanitization or filtering logic is defined for the ingested HTML content before it is processed for code generation.
Audit Metadata