israeli-media-authenticity-verifier

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local media analysis utilities including c2patool, exiftool, and ffmpeg. These calls are implemented using subprocess.run with argument lists, a secure practice that prevents shell injection via untrusted filenames.
  • [EXTERNAL_DOWNLOADS]: The instructions reference reputable verification services and tools from trusted organizations such as Google and the Content Authenticity Initiative. The skill does not perform automated downloads of unverified code.
  • [PROMPT_INJECTION]: The skill includes explicit honesty rules and safety constraints that guide the agent toward objective analysis and inconclusive verdicts when evidence is thin, rather than attempting to override model safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 08:36 PM
Security Audit — agent-trust-hub — israeli-media-authenticity-verifier