Skills
Navigate
skills/samuraigpt/generative-media-skills/muapi-selfie-with-celebrities/Gen Agent Trust Hub

muapi-selfie-with-celebrities

Pass

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill interpolates user inputs into model prompts, creating an indirect injection surface. 1. Ingestion points: movie_name and user_image in SKILL.md. 2. Boundary markers: None present. 3. Capability inventory: Uses muapi CLI and curl for network requests to api.muapi.ai. 4. Sanitization: None present for user text inputs.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands using the muapi CLI and curl. It relies on the $MUAPI_API_KEY environment variable for authenticated access to the vendor's API.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 30, 2026, 07:18 AM