keephq/keep
Keep is an open-source AIOps and alert management platform that centralizes alerts, deduplicates, enriches, filters them, and correlates incidents using AI-powered automation and various backend in...
Awesome AI for Infra › Full-Stack AIOps Platforms
databufflabs/databuff
Databuff is an AI-native Application Performance Management (APM) and Site Reliability Engineering (SRE) agent solution, built on OpenTelemetry standards. It aims to offer an open-source alternative to commercial APM tools by deeply integrating AI capabilities, rather than providing AI purely as an add-on. The platform features an AI-powered core that can directly query traces, metrics, topologies, and alerts using Large Language Models (LLMs), providing answers based on real data. It also incorporates a multi-intelligent agent collaboration framework where different AI agents (e.g., query, inspection experts) work in parallel to solve complex problems. Databuff supports AI application monitoring, including LLM call chain and token usage analysis, and plans to integrate eBPF for kernel-level, non-intrusive data collection. Its architecture is designed to be simple, using Ingest, Doris, and Web components, making it easy to deploy on Docker or Kubernetes. The system is extensible with custom skills and tools without modifying core code. It offers bidirectional openness for Metric, Call, and Peer (MCP) data, allowing integration with external MCP systems like Prometheus or SkyWalking.
https://github.com/databufflabs/databuff
Also in Full-Stack AIOps Platforms
alibaba/SREWorks
SREWorks is Alibaba Cloud's cloud-native AIOps and DataOps platform designed to enhance IT operation and maintenance through AI and big data.
William-Lu-stack/Flawless
Flawless is an AI-native SRE control plane for Kubernetes and cloud infrastructure, offering AgenticOps for incident management, remediation, and recovery verification.
alibaba/UnifiedModel
UModel is a vendor-neutral semantic runtime that provides a unified object graph for enterprise AI agents, enabling them to understand and process fragmented IT operational data, services, and busi...
nudgebee/nudgebee
Nudgebee is an open-source SRE copilot leveraging AI for observability, FinOps, runbook automation, and incident response across Kubernetes and multi-cloud environments.
HoloInsight/holoinsight
HoloInsight is a cloud-native observability platform that emphasizes real-time log analysis and integrates AI for enhanced monitoring and insights.
Kkkirito-123/Mutil-Rag-Agent
This is a multi-agent AIOps and RAG platform for on-call troubleshooting, featuring LangGraph-based diagnosis workflows, Milvus vector search, MCP tool integration, and Prometheus alerts.