cold-calling

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions define a business process for phone outreach and do not contain any prompt injection, behavior overrides, or attempts to bypass safety filters.
  • [SAFE]: The included validation script scripts/check-output.py is a benign utility that reads a local file to check for specific keywords (e.g., 'sdr', 'connect', 'calling'). It does not use unsafe functions like eval() or exec(), nor does it perform network operations or access sensitive system files.
  • [SAFE]: No obfuscation techniques, such as Base64 encoding of commands, zero-width characters, or hidden acrostics, were found in the skill content or metadata.
  • [SAFE]: The skill does not define any external dependencies or remote code download patterns. The frameworks referenced (e.g., Joey Gilkey, Ryan Reisert) are cited for informational purposes and do not involve runtime execution of untrusted code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 12:32 PM
Security Audit — agent-trust-hub — cold-calling