bascik-migrate
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to read and process external website source code (HTML, JSX, etc.) and metadata (sitemaps, robots.txt) for migration purposes. Ingestion points: external website assets and configuration files. Boundary markers: The instructions lack delimiters or warnings to ignore instructions within the ingested data. Capability inventory: Includes file system writes and command-line execution of local scripts. Sanitization: Instructions mention escaping interpolated values in output, but not for input injection.
- [COMMAND_EXECUTION]: Directs the user to run multiple local Node.js utility scripts (e.g., read-doc.mjs, verify.mjs) for auditing, documentation, and build checks.
- [EXTERNAL_DOWNLOADS]: Suggests installing the bascik-refactor skill from the vendor's repository using the platform's skill management command.
Audit Metadata