burn-graphics

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill dynamically constructs and executes ffmpeg command lines by building complex filtergraphs. It uses shell utilities to manage video files and resolve positions based on user-provided JSON parameters.
  • [PROMPT_INJECTION]: An indirect injection surface is present because the skill ingests untrusted data from an overlays JSON structure and interpolates the values directly into a shell variable (FILTER) used by ffmpeg.
  • Ingestion points: The overlays JSON input in SKILL.md.
  • Boundary markers: No explicit boundary markers or instructions are provided to delimit interpolated text strings within the shell command.
  • Capability inventory: The skill uses Bash(ffmpeg), Bash(ffprobe), Read, and Write as specified in the frontmatter of SKILL.md.
  • Sanitization: There is no evidence of sanitization, escaping, or validation logic for the user-supplied text or path fields before they are interpolated into the command string.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:34 PM
Security Audit — agent-trust-hub — burn-graphics