muapi-multi-angle-reshoot

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it interpolates user-provided text from the subject_description field directly into the generation and editing prompts without sanitization or boundary markers.
  • Ingestion points: subject_description input field in SKILL.md.
  • Boundary markers: Absent; user input is mixed directly with the prompt text used for image generation.
  • Capability inventory: The skill uses the muapi CLI and curl for subprocess calls to an external API.
  • Sanitization: None; the input is used as provided by the user.
  • [COMMAND_EXECUTION]: The skill uses the muapi CLI and curl to interact with an external image processing API. These commands are necessary for the skill's functionality and target the vendor's infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 02:05 PM
Security Audit — agent-trust-hub — muapi-multi-angle-reshoot