milestone-reveal

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Downloads the 'hyperframes' package from the heygen-com repository. HeyGen is a well-known service for AI-driven video content.
  • [REMOTE_CODE_EXECUTION]: Employs 'npx skills add' to fetch and install external skill code from heygen-com.
  • [COMMAND_EXECUTION]: Executes 'npx hyperframes render', 'brand-intake', and 'reel-export' as part of the animation and export workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill integrates user-provided data into its code generation logic. 1. Ingestion points: Milestone numbers, labels, and font files (SKILL.md). 2. Boundary markers: The prompt template uses brackets and quotes to isolate variables, though no specific instructions to ignore embedded commands are present. 3. Capability inventory: Execution of npx commands and generation of HTML/JS code (SKILL.md). 4. Sanitization: No specific input sanitization is performed on user strings.
  • [DYNAMIC_EXECUTION]: Dynamically builds an HTML/Canvas file at runtime based on user parameters and renders it using external tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 01:51 PM