azdhril/Sentinel
🛡️ Zero-trust governance for AI agents. Intercept, approve, and audit LLM actions with one decorator. Fail-secure by default.
Implements a decorator-based interception layer with JSON-configurable rules, statistical anomaly detection (Z-score analysis), and multi-channel approval workflows (terminal, webhook, dashboard). Natively integrates with LangChain tools via `protect_tools()`, and maintains JSONL audit logs for compliance; fail-secure-by-default ensures actions are blocked on any error rather than allowed.
Stars
8
Forks
1
Language
Python
License
MIT
Category
Last pushed
Jan 25, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/azdhril/Sentinel"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
TouHouQing/DataSentry
🛡️An AI-powered data governance agent platform. Supports Real-time Interception & Database Batch...
asaotomo/DeepSentry
DeepSentry（深哨） 一款 AI 驱动的新一代安全运维代理。支持本地/SSH 混合执行、动态风控与自动化审计报告生成。(集成 DeepSeek/OpenAI/Ollam/LM Stuido)
Raunplaymore/sentinel
Sentinel watches what your AI agents actually do — file access, network calls, risky commands —...
vectimus/sentinel
Automated threat intelligence pipeline for agentic AI security. Three AI agents discover...
josephawana/Kali-Safety-EDR
🛡️ Kali-Safety-EDR: A lightweight, Python-based Endpoint Detection and Response (EDR) tool...