登录

holmesgpt

HolmesGPTApache-2.0活跃维护

基于 LLM 的 SRE 智能体，自动排查 Kubernetes 等环境中的告警与故障，帮助运维团队快速定位根因、缩短事件响应时间。

SRE Agent - CNCF Sandbox Project

0.46.0发布于 2026-11-04 14:45 UTC3,852star +38 / 7 天582fork同步于 2026-11-10 23:40 UTC
SRE 智能体故障排查事件响应AIOpsKubernetes
对比 GitHub 官网aiopschatbotchatopsdevopsdevops-tools

怎么上手

由 README 提炼，执行前请确认
原文 
README 未给出具体安装命令，仅指向 https://holmesgpt.dev/installation/cli-installation/ 安装文档
硬件要求按 README 说明

README 未说明具体硬件或显存要求；Operator 模式本身需运行在 Kubernetes 中，但健康检查可查询 VM、云服务、数据库等任意已接入数据源，且不强制要求 Kubernetes。

下载：源码 zip·GitHub Releases（安装包）本站不镜像安装包

项目说明

由 README 自动翻译提炼
没人工点评？写下你的经验

CNCF 沙箱开源 SRE Agent，用 AI 自动排查生产故障并定位根因

基于 LLM 的 SRE 智能体，自动排查 Kubernetes 等环境中的告警与故障，帮助运维团队快速定位根因、缩短事件响应时间。

核心能力

  • 用 agentic loop 查询多源实时可观测数据定位根因
  • 服务端过滤与 JSON 树遍历避免大结果塞满上下文
  • 按工具限制内存并流式落盘防止查询大日志时 OOM
  • 内置 Prometheus、Grafana、Datadog、K8s 等数据源
  • 支持 AlertManager、PagerDuty 等告警双向读写

容易踩的坑

需自行配置 LLM 提供商的 API Key，README 未给出默认值；安装命令需查阅官方文档，README 未内联；接入各数据源需按文档单独配置对应 toolset。

内容由模型根据 README 生成，可能不完整或不准确，以官方文档为准。看原文 README

最近发版

5 个版本0.46.0
全部 release
  1. 0.46.0最新正式版2026-11-04 · 2026-11-04 14:45 UTC

    What's Changed

  2. 0.46.0-alphapre-release2026-11-04 · 2026-11-04 09:58 UTC
    查看更新说明

    What's Changed

  3. 0.45.02026-10-15 · 2026-10-15 08:15 UTC
    查看更新说明

    What's Changed

    Full Changelog: https://github.com/HolmesGPT/holmesgpt/compare/0.44.0...0.45.0

  4. 0.45.0-alphapre-release2026-10-15 · 2026-10-15 07:45 UTC
    查看更新说明

    What's Changed

    Full Changelog: https://github.com/HolmesGPT/holmesgpt/compare/0.44.0...0.45.0

  5. 0.44.02026-10-14 · 2026-10-14 10:19 UTC
    查看更新说明

    What's Changed

    Full Changelog: https://github.com/HolmesGPT/holmesgpt/compare/0.43.0...0.44.0

用户点评

登录用户的真实使用记录，可按版本和硬件筛选
共 0 条

    还没有人点评。第一个说出真实体验的人最有价值。

    README 原文

    未翻译
    在 GitHub 打开

    Open-source AI agent for investigating production incidents and finding root causes. Works with any stack — Kubernetes, VMs, cloud providers, databases, and SaaS platforms. We are a Cloud Native Computing Foundation sandbox project. Originally created by Robusta.Dev, with major contributions from Microsoft.

    New: Operator Mode — Find Problems 24/7 in the Background

    Most AI agents are great at troubleshooting problems, but still need a human to notice something is wrong and trigger an investigation. Operator mode fixes that — HolmesGPT runs in the background 24/7, spots problems before your customers notice, and messages you in Slack with the fix. Connect the GitHub integration and it can even open PRs to fix what it finds.

    While the operator itself runs in Kubernetes, health checks can query any data source Holmes is connected to — VMs, cloud services, databases, SaaS platforms, and more.

    Features

    • Petabyte-scale data: Server-side filtering, JSON tree traversal, and tool output transformers keep large payloads out of context windows
    • Memory-safe execution: Per-tool memory limits, streaming large results to disk, and automatic output budgeting prevent OOM kills when querying large observability datasets
    • Deep integrations: Prometheus, Grafana, Datadog, Kubernetes, and many more—plus any REST API
    • Bidirectional alert integrations: Fetch alerts from AlertManager, PagerDuty, OpsGenie, or Jira—and write findings back
    • Any LLM provider: OpenAI, Anthropic, Azure, Bedrock, Gemini, and more
    • No Kubernetes required: Works with any infrastructure — VMs, bare metal, cloud services, or containers

    How it Works

    HolmesGPT uses an agentic loop to query live observability data from multiple sources and identify root causes.

    HolmesGPT Investigation Demo

    🔗 Data Sources

    HolmesGPT integrates with popular observability and cloud platforms. The following data sources ("toolsets") are built-in. Add your own.

    Data SourceNotes
    AKSAzure Kubernetes Service cluster and node health diagnostics
    Atlassian RovoJira issues and Confluence pages via Atlassian's hosted server (MCP)
    ArgoCDGet status, history and manifests and more of apps, projects and clusters
    AWSAny AWS API via the hosted AWS MCP Server: RDS, EC2, CloudWatch, CloudTrail, Cost Explorer and more (MCP)
    AzureAzure resources and diagnostics (MCP)
    ConfluencePrivate runbooks and documentation
    Confluence (MCP)Private runbooks and documentation (MCP)
    CoralogixRetrieve logs for any resource
    CrossplaneTroubleshoot Crossplane providers, compositions, claims, and managed resources
    DatadogQuery logs, metrics, and traces
    DockerGet images, logs, events, history and more
    Elasticsearch / OpenSearchQuery logs, cluster health, shard and index diagnostics
    GCPGoogle Cloud Platform resources (MCP)
    GitHubRepositories, issues, and pull requests (MCP)
    GitLabProjects, merge requests, issues, and CI/CD pipelines (MCP)
    Jenkins (MCP)Build status, pipeline logs, and job history (MCP)
    GrafanaQuery and analyze dashboard configurations and panels
    HelmRelease status, chart metadata, and values
    InternetPublic runbooks, community docs, etc.
    KafkaFetch metadata, list consumers and topics or find lagging consumer groups
    KubernetesPod logs, K8s events, and resource status (kubectl describe)
    Kubernetes Remediation (MCP)Apply fixes like scaling, rollbacks, and resource edits (MCP)
    LokiQuery logs for Kubernetes resources or any query
    MariaDBMariaDB database queries and diagnostics
    MongoDBQuery data, diagnose performance, inspect schemas, find slow operations
    MongoDB AtlasCluster health, slow queries, and performance diagnostics
    NewRelicInvestigate alerts, query tracing data
    OpenShiftProjects, routes, builds, security context constraints, and deployment configs
    Prefect (MCP)Workflow orchestration monitoring, flow runs, and worker health (MCP)
    PrometheusInvestigate alerts, query metrics and generate PromQL queries
    RabbitMQPartitions, memory/disk alerts, troubleshoot split-brain scenarios and more
    RobustaMulti-cluster monitoring, historical change data, runbooks, PromQL graphs and more
    ServiceNowQuery tables and incident records
    SentryError tracking, issues, and performance monitoring (MCP)
    SlabTeam knowledge base and runbooks on demand
    SplunkLog search and analysis (MCP)
    SQL DatabasesPostgreSQL, MySQL, ClickHouse, MariaDB, SQL Server, Azure SQL, SQLite
    TempoFetch trace info, debug issues like high latency in application
    VictoriaLogsQuery logs from VictoriaLogs using LogsQL
    VictoriaMetricsQuery metrics from a Prometheus-compatible TSDB (vmsingle / vmcluster)
    ZabbixMonitor hosts, problems, events, triggers, and historical metrics

    See the full list of built-in toolsets for additional integrations including Cilium, KubeVela, Notion, and more.

    🚀 End-to-End Automation

    HolmesGPT can fetch alerts/tickets to investigate from external systems, then write the analysis back to the source or Slack.

    IntegrationStatusNotes
    Slack✅Demo. Available via Robusta
    Microsoft Teams✅Available via Robusta
    Prometheus/AlertManager✅Robusta or HolmesGPT CLI
    PagerDuty✅HolmesGPT CLI only
    OpsGenie✅HolmesGPT CLI only
    Jira✅HolmesGPT CLI only
    GitHub✅HolmesGPT CLI only

    Installation

    Read the installation documentation to learn how to install HolmesGPT.

    Supported LLM Providers

    Read the LLM Providers documentation to learn how to set up your LLM API key.

    Using HolmesGPT

    See the walkthrough documentation for usage guides, including:

    🔐 Data Privacy

    By design, HolmesGPT has read-only access and respects RBAC permissions. It is safe to run in production environments.

    License

    Distributed under the Apache 2.0 License. See LICENSE for more information.

    Community

    Join our community to discuss the HolmesGPT roadmap and share feedback:

    Support

    If you have any questions, feel free to message us on HolmesGPT Slack Channel

    How to Contribute

    Please read our CONTRIBUTING.md for guidelines and instructions.

    For help, contact us on Slack or ask DeepWiki AI your questions.

    Please make sure to follow the CNCF code of conduct - details here. Ask DeepWiki

    OpenSSF Best Practices OpenSSF Scorecard

    相关项目

    同分类里 star 最高的项目
    对比全部 →
    radare2
    radareorg·C

    类 UNIX 的逆向工程框架和命令行工具集，用于反汇编、二进制分析与恶意软件取证，支持脚本化操作。

    逆向工程+4恶意软件分析 / 取证
    25388+353.4k6.2.62026-11-10 23:16 UTC
    witr
    pranshuparmar·Go

    一个用 Go 写的命令行工具，能把任意进程、端口、容器或文件反向追溯到是谁启动的，排查线上问题时不用再靠猜。

    进程溯源+4恶意软件分析 / 取证
    22978+23810v0.3.42026-10-23 03:02 UTC
    kubeshark
    kubeshark·Go

    基于 eBPF 的 Kubernetes 网络可观测性工具，能索引带完整 K8s 上下文的 L4/L7 流量，无需密钥即可解密 TLS，支持通过 MCP 供 AI 代理查询或看板人工排查。

    网络可观测性+4恶意软件分析 / 取证
    12135+2556v53.6.02026-11-10 16:52 UTC
    Detect-It-Easy
    horsicq·JavaScript

    跨平台文件类型识别工具，通过签名、熵值和结构分析判断二进制文件格式，常用于恶意软件分析与逆向工程。

    文件类型识别+4恶意软件分析 / 取证
    11935+34998db_extra2026-11-10 06:43 UTC