DevOps AI Toolkit

AI-powered platform engineering and DevOps automation through intelligent Kubernetes operations and conversational workflows.

AI Engine Docs | MCP Setup

Overview

DevOps AI Toolkit brings AI-powered intelligence to platform engineering, Kubernetes operations, and development workflows. Access it through MCP for AI coding assistants or the CLI for direct agent integration.

Key capabilities:

• Intelligent Kubernetes deployment recommendations
• AI-powered issue remediation and root cause analysis
• Semantic search over organizational documentation
• Shared prompt libraries for consistent workflows

Recommended configuration: use AI_PROVIDER=deepseek for the main AI model and EMBEDDINGS_PROVIDER=openai with OPENAI_API_KEY for embeddings.

Plugin discovery supports both Kubernetes and local environments: mount plugins.json at /etc/dot-ai/plugins.json, or set DOT_AI_PLUGINS_CONFIG to the same JSON array when running outside the cluster.

Local Development

Use dot.nu to manage a local Kind cluster for development:

# Full setup — builds local image, creates Kind cluster, deploys everything
./dot.nu setup

# Setup without building local image (use published image instead)
./dot.nu setup --build-local false

# Setup with a specific stack version
./dot.nu setup --stack-version 0.45.0 --kyverno-enabled false --atlas-enabled false

# Rebuild and redeploy after code changes (no cluster recreation)
./dot.nu redeploy

# Deploy IC-MS/ID-MS stubs for e2e integration testing
./dot.nu deploy stubs --kubeconfig kubeconfig.yaml

# Tear down the cluster
./dot.nu destroy

API key precedence:

• LLM: AI_PROVIDER (if set) always wins, so ANTHROPIC_API_KEY can be left set in your shell for another tool without hijacking this app's provider. If AI_PROVIDER is unset, ANTHROPIC_API_KEY → Anthropic (default, Claude Haiku) is still the zero-config quick-start path. Otherwise set AI_PROVIDER + AI_PROVIDER_KEY + AI_PROVIDER_BASE_URL + AI_MODEL.
• Embeddings: OPENAI_API_KEY → native OpenAI. Otherwise set EMBEDDINGS_API_KEY + EMBEDDINGS_BASE_URL + EMBEDDINGS_MODEL + EMBEDDINGS_DIMENSIONS.

Keys are read from environment variables — no interactive prompts.

After setup completes, all non-empty provider vars are saved to .env. Run source .env to load them into your shell. Variables saved:

| Variable | When saved | |---|---| | DOT_AI_AUTH_TOKEN | always | | AI_PROVIDER | always | | AI_MODEL | when set | | AI_PROVIDER_KEY | when set (custom provider only) | | AI_PROVIDER_BASE_URL | when set (custom provider only) | | EMBEDDINGS_API_KEY | when set (custom embeddings only) | | EMBEDDINGS_BASE_URL | when set (custom embeddings only) | | EMBEDDINGS_MODEL | when set (custom embeddings only) | | EMBEDDINGS_DIMENSIONS | when set (custom embeddings only) |

ANTHROPIC_API_KEY and OPENAI_API_KEY are saved separately by their own setup steps.

Deployment

For the easiest setup, we recommend installing the complete dot-ai stack which includes all components pre-configured. See the Stack Installation Guide.

For individual component installation, see the Deployment Guide.

AI Engine Docs | MCP Setup

Support

• Support Guide - How to get help and where to ask questions
• GitHub Issues: Bug reports and feature requests
• GitHub Discussions: Community Q&A and discussions

Contributing & Governance

We welcome contributions from the community! Please review:

• Contributing Guidelines - How to contribute code, docs, and ideas
• Code of Conduct - Community standards and expectations
• Security Policy - How to report security vulnerabilities
• Governance - Project governance and decision-making
• Maintainers - Current project maintainers
• Roadmap - Project direction and priorities

License

MIT License - see LICENSE file for details.

Telemetry

This project collects anonymous usage analytics to improve the product. Learn more or opt out.

Acknowledgments

DevOps AI Toolkit is built on:

• Model Context Protocol for AI integration framework
• Vercel AI SDK for unified AI provider interface
• Kubernetes for the cloud native foundation
• CNCF for the cloud native ecosystem

DevOps AI Toolkit - Making cloud native operations accessible through AI-powered intelligence.