remotion-video-composition

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a structured, production-oriented workflow for React-based video composition using the Remotion framework. It provides specific guidance on project architecture, asset management, and quality assurance.\n- [INDIRECT_PROMPT_INJECTION]: The skill involves processing external data (JSON files, media assets, and input props) to drive video generation. This represents an attack surface for indirect prompt injection if the data contains malicious instructions. However, the skill explicitly mitigates this risk by requiring Zod schema validation (projectSchema, variantSchema) for all input props and metadata before rendering.\n
  • Ingestion points: Data enters the agent context through src/data/*.json files, assets in public/assets/, and user-provided props processed via calculateMetadata.\n
  • Boundary markers: The skill demonstrates and mandates the use of Zod schemas to define strict boundaries for valid data.\n
  • Capability inventory: Capabilities are limited to deterministic React rendering and standard media processing via Remotion and FFmpeg. No arbitrary command execution or network calls at render-time are permitted.\n
  • Sanitization: Input is sanitized through schema validation and the requirement that all inputs be JSON-serializable.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 06:41 PM