web-development
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the CloudBase Web SDK from the vendor's official CDN at
https://static.cloudbase.net/cloudbase-js-sdk/latest/cloudbase.full.jsfor use in static web pages. - [COMMAND_EXECUTION]: The skill requires the execution of standard build and linting commands, including
npm run build,vite build,tsc --noEmit, andeslint, as part of its mandatory self-verification process. - [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest untrusted data from web pages via a browser tool, which could be used for indirect prompt injection attacks.
- Ingestion points: The
agent-browsertool is used to access local dev servers or external sites to reproduce bugs and verify features (specified inbrowser-testing.md). - Boundary markers: None identified; no specific guidance is given to distinguish between instructions and data when interacting with browser content.
- Capability inventory: The skill has capabilities for file system modification, command execution, and cloud platform management using tools like
manageHostingandmanageFunctions(specified inSKILL.mdandframeworks.md). - Sanitization: No validation or sanitization requirements are mentioned for the content retrieved through the browser validation flow.
Audit Metadata