skills/duc01226/easyplatform/remotion/Gen Agent Trust Hub

remotion

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains directives to "Ignore Claude-specific mode-switch instructions" and enforces a "Strict execution contract." These instructions are intended to stabilize the agent's behavior during complex video orchestration tasks.
  • [COMMAND_EXECUTION]: Shell commands are used for project lifecycle management, including initialization via npx create-video, dependency installation with npm install, and starting the local development server with npm run studio.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes npx to execute tools like create-video and remotion CLI directly, which involves running remote code from the official Remotion registry.
  • [EXTERNAL_DOWNLOADS]: Integrations with well-known services such as ElevenLabs for text-to-speech, Mapbox for map rendering, and Google Fonts for typography are documented. These interactions are standard for the skill's purpose.
  • [CREDENTIALS_SAFE]: The skill identifies the need for API keys for external services but appropriately advises users to manage these secrets through .env files, avoiding hardcoded credentials in the codebase.
  • [INDIRECT_PROMPT_INJECTION]: As the skill processes user-defined video descriptions to generate code, it possesses an ingestion surface for external data. However, the capabilities are scoped to project-specific files and standard development tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 03:12 PM
Security Audit — agent-trust-hub — remotion