# typesafe_ai v0.1.1 - Table of Contents

> An Elixir client for TypeSafe AI with typed responses and bounded concurrency

## Pages

- Start here
  - [Overview](readme.md)
  - [Getting started](getting-started.md)

- Guides
  - [Configuration and concurrency](configuration.md)
  - [Errors and retries](errors-and-retries.md)
  - [Telemetry](telemetry.md)
  - [Performance and capacity](performance.md)
  - [Support triage and evaluation](examples.md)

- Releases
  - [Changelog](changelog.md)
  - [License](license.md)

## Modules

- Client API
  - [TypeSafe](TypeSafe.md): An Elixir client for the TypeSafe AI System One API.
  - [TypeSafe.Client](TypeSafe.Client.md): A supervised TypeSafe AI client with bounded concurrency and queues.
  - [TypeSafe.Question](TypeSafe.Question.md): A question about the request's shared state.

- Results
  - [TypeSafe.Answer.Choice](TypeSafe.Answer.Choice.md): The answer to a question built with `TypeSafe.choice/2`.
  - [TypeSafe.Answer.Noul](TypeSafe.Answer.Noul.md): The answer to a question built with `TypeSafe.noul/2`.
  - [TypeSafe.Answer.Score](TypeSafe.Answer.Score.md): The answer to a question built with `TypeSafe.score/2`.
  - [TypeSafe.Response](TypeSafe.Response.md): A successful evaluation returned by `TypeSafe.system_one/2`.

- Errors and retries
  - [TypeSafe.Error](TypeSafe.Error.md): A failure returned as `{:error, error}` by the client.
  - [TypeSafe.Retry](TypeSafe.Retry.md): Bounded retry policy. `max_attempts` includes the initial request.

