rebuild-site
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to execute local Python scripts (mvsetup.py,ingest.py,assetgen.py) and theffmpegcommand-line utility for image processing. These executions are constrained to local scripts and media assets. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data (video frames and a generated analysis report) to identify layout and transcribe copy, which introduces an indirect prompt injection surface.
- Ingestion points: Video frames and the
report.mdfile containing timeline and layout metadata. - Boundary markers: No specific boundary markers or 'ignore' instructions are used when processing the transcribed copy.
- Capability inventory: The agent has access to
Bash,Read,Write, andEdittools. - Sanitization: No explicit sanitization of external content is performed before interpolation into the site plan or code generation.
Audit Metadata