kubectl-ai is an AI-powered Kubernetes assistant written in Go that translates natural language user intent into precise Kubernetes operations. The tool acts as an intelligent interface designed to make Kubernetes management more accessible and efficient by allowing users to interact with their clusters through conversational queries rather than memorizing complex kubectl commands.
The project supports multiple AI model providers, giving users flexibility in choosing their backend. Users can leverage Google's Gemini API as the default option, or alternatively use models from Vertex AI, Azure OpenAI, OpenAI, Grok, AWS Bedrock, and local LLM providers such as ollama and llama.cpp. This multi-provider approach allows teams to integrate kubectl-ai with their existing AI infrastructure and preferences. The tool can be installed via quick install scripts on Linux and macOS, through manual download from the releases page, via krew the kubectl plugin manager, or on NixOS through package management.
kubectl-ai operates in multiple modes to suit different workflows. Users can run it interactively to maintain conversational context across multiple queries, execute single tasks from the command line, or pipe input from other Unix commands. The tool supports session persistence, allowing users to save and resume conversations to maintain context between different runs and interfaces. Configuration is flexible, supporting YAML configuration files at ~/.config/kubectl-ai/config.yaml, command-line flags, and environment variables, with command-line flags taking precedence.
The tool extends its capabilities through a tools system that leverages built-in kubectl and bash tools while allowing users to define custom tools through configuration files. kubectl-ai also supports the Model Context Protocol, enabling it to function as both an MCP client that connects to external MCP servers for additional tools and as an MCP server that exposes Kubernetes tools to other MCP clients like Claude, Cursor, or VS Code. The enhanced MCP server mode creates a tool aggregation hub where kubectl-ai can consume tools from other servers while exposing its own Kubernetes capabilities.
According to GitGenius activity tracking, the repository shows strong engagement with a median issue and pull request response latency of 0.0 hours and a mean latency of 19.0 hours across 169 tracked items. The most active labels are enhancement with 63 items, bug with 44 items, and help wanted with 34 items. Primary contributors include droot with 338 tracked events, noahlwest with 65 events, and tuannvm with 57 events. The repository shares contributors with major projects including Microsoft's VS Code and TypeScript repositories as well as the Rust language repository, indicating cross-pollination with significant open source ecosystems.
The project provides Docker support for standalone environments and includes a modelserving directory with tools for deploying llama.cpp-based LLM serving endpoints locally or on Kubernetes clusters. kubectl-ai can be invoked as a standard kubectl plugin using kubectl ai, integrating seamlessly into existing kubectl workflows. The repository welcomes community contributions and provides comprehensive documentation including contribution guidelines and learning resources such as talks on the MCP architecture and practical usage patterns.
