video-canvas-expand

Warn

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a shell command using string interpolation for user input: meitu video-canvas-expand ... --prompt "{扩展内容描述}". This pattern is vulnerable to command injection if the input contains shell metacharacters (e.g., ;, &, |, or $()), allowing an attacker to execute arbitrary system commands.
  • [CREDENTIALS_UNSAFE]: The skill configuration explicitly requests read access to sensitive credential files located at ~/.meitu/credentials.json. While these are likely intended for the meitu tool's authentication, the availability of these secrets to the agent environment poses a risk of credential exposure if the agent is compromised or misled.
  • [COMMAND_EXECUTION]: The skill requires the execution of a local binary meitu. While associated with a known vendor, the skill's implementation relies on the presence of this external executable without verifying its integrity or origin within the instruction file.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 15, 2026, 06:03 AM
Security Audit — agent-trust-hub — video-canvas-expand