Skills
skills/tangyang/skills/meitu-image-edit/Gen Agent Trust Hub

meitu-image-edit

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script run_command.py to perform the image-edit task. This follows a standard and secure pattern for tool integration using a fixed command string and parameters passed via a JSON object.\n- [PROMPT_INJECTION]: The skill accepts a prompt input for image editing, creating a surface for indirect prompt injection where user data is processed by the agent.\n
  • Ingestion points: prompt field in the input contract within SKILL.md.\n
  • Boundary markers: None identified in the skill definition.\n
  • Capability inventory: Execution of the image-edit command via a local wrapper script as defined in SKILL.md.\n
  • Sanitization: No specific sanitization or filtering logic for the prompt input is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 08:48 AM