Skip to content
    Back to AIOps & Observability
    AIOps & ObservabilityStartupCNCF Sandbox

    K8sGPT

    CNCF Sandbox project that scans Kubernetes clusters and uses LLMs to explain failures in plain English and suggest fixes — CLI or continuous in-cluster operator

    Mkt Cap / ValOpen Source
    May 2026: v0.4.33 ships CNCF incubation prep
    De facto open-source standard for AI-assisted Kubernetes diagnostics, under CNCF governance.
    Analyst take · Competitive edge

    SWOT Analysis

    Strengths
    • CNCF Sandbox project with active, vendor-neutral governance
    • 8k GitHub stars and 1k forks with steady release cadence into 2026
    • Supports many AI backends including OpenAI, Bedrock, Azure, and local models
    • Operator mode enables continuous in-cluster scanning, not just one-off CLI runs
    • Free and open source, easy bottom-up adoption by platform teams
    Opportunities
    • CNCF incubation path underway per May 2026 governance prep
    • Local model support for privacy-sensitive enterprises
    • Integration into platform engineering golden paths
    • Agent and MCP ecosystem integrations
    Weaknesses
    • No commercial support or SLA behind the project
    • Diagnostic quality depends on the chosen LLM backend
    • Scope limited to Kubernetes diagnostics, not full observability
    • Still Sandbox maturity; incubation not yet achieved
    Threats
    • AI assistants built into kubectl and hyperscaler consoles
    • Observability vendors bundling K8s AI troubleshooting
    • General-purpose coding agents handling cluster debugging
    • Maintainer and contributor sustainability risk

    User Sentiment

    Synthesized from G2, Gartner Peer Insights, and analyst review data.

    What users love
    • Plain-English explanations of cryptic cluster errors
    • Fast triage that saves SRE investigation time
    • Simple install as CLI or Kubernetes operator
    • Built-in anonymization of sensitive cluster data
    Common complaints
    • LLM token costs add up on large clusters
    • Remediation advice can be generic for complex failures
    • Configuring AI backends and auth can be fiddly

    Customer Profile

    Who buys this

    Typical segments

    Platform engineering teamsKubernetes-heavy enterprises

    Typical buyer

    Platform engineering lead or SRE manager adopting bottom-up via OSS

    Top use cases
    1. 1Triaging failing pods and misconfigurations
    2. 2Continuous in-cluster scanning via operator
    3. 3Helping junior engineers debug Kubernetes

    Future Focus Areas

    1

    CNCF incubation and graduated maturity

    2

    Deeper automated remediation workflows

    3

    Expanded local and open-model backends

    4

    Tighter observability stack integrations