video-stitch

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the meitu CLI tool to perform video processing tasks. It constructs shell commands by interpolating user-provided inputs, such as video_list and music_url, directly into command-line arguments.
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection and command injection due to the lack of explicit sanitization for user-provided data used in shell execution.
  • Ingestion points: The skill accepts video_list, music_url, and prompt parameters from the user in SKILL.md.
  • Boundary markers: There are no defined boundary markers or instructions to ignore embedded commands within the interpolated parameters.
  • Capability inventory: The skill has the capability to execute the meitu binary and perform filesystem read/write operations in the ~/.meitu/ and workspace directories.
  • Sanitization: No sanitization, escaping, or validation logic for the input URLs or prompt strings is mentioned in the skill definition.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 06:03 AM
Security Audit — agent-trust-hub — video-stitch