tgravity-work-mcn

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No prompt injection or bypass patterns detected. The instructions are focused on routing logic and intent classification.
  • [SAFE]: No sensitive data exposure or exfiltration risks. The skill explicitly forbids the agent from storing or displaying sensitive documents like contracts, seals, or private contact info.
  • [SAFE]: No obfuscation, hidden characters, or encoded content found in the instructions or metadata.
  • [SAFE]: No external dependencies, remote code execution, or package installations are utilized.
  • [SAFE]: The skill does not request or attempt to acquire elevated privileges (e.g., sudo, chmod).
  • [SAFE]: Indirect prompt injection risk is minimal as the skill has no tool-calling capabilities and explicitly instructs the agent not to save, index, or create persistent records from user input.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 08:00 AM
Security Audit — agent-trust-hub — tgravity-work-mcn