multiview-fit-loop

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security concerns were identified in the skill instructions or the accompanying Python script.
  • [EXTERNAL_DOWNLOADS]: The skill uses standard image processing libraries (OpenCV and NumPy) which are typical for its stated purpose of visual template matching.
  • [DATA_EXFILTRATION]: All data processing is performed locally on the user's filesystem. There are no network requests or attempts to exfiltrate data.
  • [PROMPT_INJECTION]: The instructions focus purely on the technical logic of the fitting loop and do not attempt to override agent safety protocols or bypass constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 08:28 PM
Security Audit — agent-trust-hub — multiview-fit-loop