Search

Search docs and pages

Pages

Athas Agent

Athas Agent is the built-in AI chat surface in Athas. Use it when you want to ask questions, reason through code, or work with files you reference from the editor.

Common Uses

  • Ask questions about a file, function, error, or project structure
  • Discuss an implementation plan before editing code
  • Generate code snippets, tests, refactors, or explanations
  • Switch models when a task needs a different provider or capability

Open the Agent

Open the AI panel from the sidebar or with Cmd+Shift+A on macOS and Ctrl+Shift+A on Linux and Windows.

Configure the Model

Go to Settings > AI and choose:

  • Provider: the service Athas Agent should use
  • Model: the model sent to that provider
  • Temperature: lower values are more deterministic, higher values are more creative
  • Max Tokens: the maximum response length

Custom Provider

Select Custom when you want Athas Agent to use an OpenAI-compatible endpoint.

You need:

  • A base URL, such as https://api.example.com/v1
  • A model ID exactly as your provider expects it
  • An optional API key if the endpoint requires a bearer token

Athas can accept either the API root or a full chat completions URL. For example, https://api.example.com/v1 and https://api.example.com/v1/chat/completions are both valid.

See Providers and Models for provider setup.