Kubently
Troubleshoots Kubernetes clusters agentically, diagnosing issues via natural-language queries through the ask_kubently tool.
Kubently is an AI agent that debugs Kubernetes clusters by running read-only kubectl commands in response to natural-language questions, exposed through the ask_kubently tool. It supports both the MCP and A2A protocols so any MCP client or agent-to-agent system can connect, and it works with multiple LLM providers including Anthropic, OpenAI, and Google Gemini. Deployments run inside a user's own cluster, with API-key or OAuth/OIDC authentication.
Source
Repository: https://github.com/kubently/kubently
Maintain Kubently?
Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.
[Kubently on getagentictools](https://getagentictools.com/mcp/kubently-kubently?ref=badge) Related mcp
Kong Konnect Integrates with Kong API Gateway to enable management and analysis of API configurations, traffic patterns, and control plane sett
pyResToolbox Provides reservoir engineering calculations through the pyResToolbox library, exposing 47 tools for oil and gas PVT properties, in
PowerPlatform Dataverse Integrates with Microsoft PowerPlatform/Dataverse to enable intelligent access to entity metadata, attributes, relationships, and
Coolify Enables comprehensive Coolify infrastructure management by exposing tools for creating, deploying, and tracking servers, applicati
Excalidraw Enables creation and management of Excalidraw drawings with tools for creating, updating, exporting, and storing visual content di
Kubernetes Integrates with Kubernetes clusters to provide resource management, monitoring, and troubleshooting capabilities across multiple c