convert-video

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute system commands including ffmpeg and ffprobe. It constructs these commands by interpolating variables like $INPUT, $OUTPUT, and $N which are derived from user-provided data. This pattern can lead to command injection if the agent does not properly escape shell metacharacters in file paths or parameters.- [PROMPT_INJECTION]: The skill ingests untrusted data from user requests and file system metadata which creates an indirect prompt injection surface.
  • Ingestion points: User-provided video file paths and transformation parameters (e.g., seek times, scale dimensions, and frame rate variables).
  • Boundary markers: Absent. There are no explicit delimiters or instructions to the agent to treat user-provided strings as data rather than executable parts of the command.
  • Capability inventory: The skill utilizes Bash(ffprobe:*) and Bash(ffmpeg:*) which allow execution of powerful command-line utilities.
  • Sanitization: Absent. The skill provides command templates but lacks instructions for input validation, type checking, or shell escaping of the variables used in the CLI invocations.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:40 PM