Files
hermes-agent/agent/shield.py
Google AI Agent 3d8cf5122a
Some checks failed
Contributor Attribution Check / check-attribution (pull_request) Failing after 31s
Docker Build and Publish / build-and-push (pull_request) Has been skipped
Supply Chain Audit / Scan PR for supply chain risks (pull_request) Successful in 40s
Tests / e2e (pull_request) Successful in 2m2s
Tests / test (pull_request) Failing after 52m0s
feat: add agent/shield.py for SHIELD defense
2026-04-20 15:54:48 +00:00

25 lines
713 B
Python

import logging
from tools.shield.detector import ShieldDetector, Verdict, CRISIS_SYSTEM_PROMPT, SAFE_SIX_MODELS
logger = logging.getLogger(__name__)
_detector = None
def get_detector():
global _detector
if _detector is None:
_detector = ShieldDetector()
return _detector
def scan_text(text: str):
"""Scan text for jailbreaks and crisis signals using SHIELD."""
detector = get_detector()
return detector.detect(text)
def is_crisis(verdict: str) -> bool:
return verdict in [Verdict.CRISIS_DETECTED.value, Verdict.CRISIS_UNDER_ATTACK.value]
def is_jailbreak(verdict: str) -> bool:
return verdict in [Verdict.JAILBREAK_DETECTED.value, Verdict.CRISIS_UNDER_ATTACK.value]