happy-horse

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits user-provided media files and an API key to the external domain ai-hive.iclip.cn. It also uses the webbrowser module to open external URLs for user onboarding.
  • [EXTERNAL_DOWNLOADS]: The script downloads generated video files from remote result URLs provided by the AI Hive API to the local file system.
  • [PROMPT_INJECTION]: The skill ingests untrusted prompt strings and media from users to be processed by an external AI service, creating a surface for indirect prompt injection attacks. 1. Ingestion points: CLI arguments --prompt, --image, --video, --audio, --first-frame, and --last-frame in scripts/videogen.py. 2. Boundary markers: None detected; prompts are interpolated directly into API request bodies. 3. Capability inventory: File system read/write access and network communication with ai-hive.iclip.cn. 4. Sanitization: None; input strings are not sanitized or escaped before transmission.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 12:20 PM
Security Audit — agent-trust-hub — happy-horse