video-pro
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses nodes.run to execute shell commands such as ffmpeg and npx remotion on a target MiniPC node to perform video editing and rendering tasks.
- [EXTERNAL_DOWNLOADS]: The skill uses npx to run Remotion, which may result in fetching packages from the public npm registry during execution.
- [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection by instructing the agent to interpolate data (such as composition IDs, video titles in props, and file paths) directly into shell commands. This could allow a malicious user to escape the command structure and execute arbitrary commands on the MiniPC node. 1. Ingestion points: Remotion props JSON, composition IDs, and file names mentioned in SKILL.md. 2. Boundary markers: No delimiters or ignore instructions are present for the interpolated data. 3. Capability inventory: Full shell command execution via nodes.run on the remote node. 4. Sanitization: No evidence of input validation or escaping is provided in the instructions.
Audit Metadata