remotion

Warn

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to clone a repository from an untrusted GitHub account (https://github.com/Ceeon/remotion-skill.git) if the local environment is missing.
  • [REMOTE_CODE_EXECUTION]: After cloning the repository, the skill runs npm install (which can execute arbitrary install scripts) and remotion render. Executing code from an unverified source is a significant security risk.
  • [COMMAND_EXECUTION]: The skill uses various shell commands and scripts: npm install for dependency management, git clone for code acquisition, and npx remotion render for execution. It also runs a local environment check script 脚本库/check-env.sh.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by converting user-provided natural language or ASCII sketches into executable React code.
  • Ingestion points: User descriptions and sketches in the '理解用户需求' step.
  • Boundary markers: No delimiters or safety instructions are defined for processing user input.
  • Capability inventory: Access to git, npm, and remotion CLI tools.
  • Sanitization: No sanitization logic for the generated code is provided before rendering.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 21, 2026, 01:42 PM
Security Audit — agent-trust-hub — remotion