video-effect-apply

Warn

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFEDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the meitu command-line utility by interpolating user-provided inputs like {url} and {特效描述} directly into shell command strings.
  • Evidence: The execution logic in SKILL.md shows commands such as meitu video-effect-apply --video_url {url} --prompt "{特效描述}" --json ....
  • [CREDENTIALS_UNSAFE]: The skill is designed to read sensitive API credentials from a local JSON file.
  • Evidence: The YAML frontmatter and requirements section specify access to ~/.meitu/credentials.json.
  • [DATA_EXFILTRATION]: The skill requests broad read permissions for local configuration and credential directories, which could be exposed if the tool's execution context is compromised.
  • Evidence: Permission to read ~/.meitu/credentials.json and ~/.meitu/tool-registry.json is explicitly requested in the metadata.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 15, 2026, 06:03 AM
Security Audit — agent-trust-hub — video-effect-apply