remotion-expert

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION] (LOW): The skill demonstrates network operations to external, non-whitelisted domains to retrieve video metadata and assets.
  • Evidence: fetch(https://api.v2.com/video-data/${props.id}`)` in SKILL.md and references/compositions-props.md.
  • Note: These operations are typical for dynamic video workflows but represent a data egress surface.
  • [PROMPT_INJECTION] (LOW): The skill identifies surfaces for indirect prompt injection via dynamic props and natural language handlers.
  • Ingestion points: External data fetched in calculateMetadata and natural language instructions passed to aiUpdateHandler.
  • Boundary markers: Absent; the code patterns do not include delimiters or 'ignore embedded instructions' warnings when processing external data.
  • Capability inventory: The skill utilizes fetch for network access and @remotion/renderer for code execution during the video rendering process.
  • Sanitization: No evidence of input validation or sanitization is present in the provided patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:03 PM