# atmos ai

Atmos AI is a built-in assistant that understands your stacks, components, and configuration.
Ask questions, debug issues, or automate infrastructure tasks directly from the terminal.
It supports API providers (Anthropic, OpenAI, Gemini, Ollama, AWS Bedrock, Azure OpenAI) and
CLI providers (Claude Code, OpenAI Codex, Gemini CLI) that reuse your existing subscription.
Choose from [21+ specialized skills](/cli/configuration/ai/skills) and switch between them with `Ctrl+A`.

> ⚠️ Experimental

**Configure AI**

Learn how to configure AI providers, skills, tools, and sessions in your atmos.yaml.

Configuration Reference[Read more](/cli/configuration/ai)

## Global `--ai` Flag

You can also add `--ai` to **any** Atmos command to get AI-powered analysis of the output — no need
to use `atmos ai` subcommands. Pair with `--skill` for domain-specific expertise
(the `--skill` flag requires `--ai`). Multiple skills can be combined with commas or repeated flags.

```shell
# AI analyzes the terraform plan output
atmos terraform plan vpc -s ue1-prod --ai

# AI uses Terraform expertise for deeper analysis
atmos terraform plan vpc -s ue1-prod --ai --skill atmos-terraform

# Combine multiple skills (comma-separated)
atmos terraform plan vpc -s ue1-prod --ai --skill atmos-terraform,atmos-stacks

# Combine multiple skills (repeated flag)
atmos terraform plan vpc -s ue1-prod --ai --skill atmos-terraform --skill atmos-stacks

# Enable via environment variables
ATMOS_AI=true ATMOS_SKILL=atmos-terraform,atmos-stacks atmos terraform plan vpc -s ue1-prod
```

See [Global Flags](/cli/global-flags) for details.

## Subcommands