Skip to main content

Welcome to Sentinel AI

Sentinel AI is an autonomous DevOps agent designed to monitor, diagnose, and repair problems in Linux servers in real-time. Using Generative AI (LLMs) powered by a RAG (Retrieval-Augmented Generation) system, it makes informed decisions based on official technical documentation.

Quickstart

Get Sentinel AI running in minutes

Agent Workflow

Understand how the autonomous agent works

API Reference

Explore the complete API documentation

Dashboard

Real-time monitoring and control interface

Key Features

Autonomous Monitoring

Continuously monitors Linux services including Nginx, PostgreSQL, Docker, and SSH

RAG-Powered Diagnosis

Uses Pinecone, LlamaIndex, and Cohere Rerank to query official technical documentation

LangGraph Workflow

Intelligent agent workflow with human-in-the-loop approval for critical actions

Real-Time Dashboard

Modern Next.js interface with live terminal and WebSocket updates

Secure Execution

SSH-based command execution with safety controls and approval workflows

Interactive Chat

Query technical documentation through an AI-powered chat interface

How It Works

The agent operates through an intelligent decision graph:
  1. Monitor — Detects failed services or anomalies
  2. Diagnose — Analyzes logs and errors using RAG to find root causes
  3. Plan — Generates step-by-step repair plans
  4. Approve — Requests human approval for critical commands
  5. Execute — Executes commands via SSH securely
  6. Verify — Confirms the issue is resolved

Get Started

Follow the quickstart guide to set up Sentinel AI

Configuration

Configure API keys, SSH, and services

Build docs developers (and LLMs) love