> ## Documentation Index
> Fetch the complete documentation index at: https://docs.moonbounce.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Score content against a decision you describe in code, and teach it with feedback.

The Moonbounce Decisions API scores content against a **decision** you describe in code. A decision is a small JSON spec of indicators (**features**) an engine detects in your content, plus an **aggregator** that turns the answers into one score. Every evaluation is recorded, and the feedback you send back is what previews and suggested improvements learn from.

Your code owns the decision and the cut points that turn a score into an outcome such as `approve` or `escalate`. Moonbounce runs the engine, stores the answers and shows you how a change would have done on past cases.

> [!WARNING]
>
> This isn't a public API, and it isn't stable yet. Access is by invitation: join the waitlist at [app.moonbounce.ai](https://app.moonbounce.ai), and once you're granted access you can sign up, create your team and create an API key on its **Keys** page ([app.moonbounce.ai/team/keys](https://app.moonbounce.ai/team/keys)). Names and shapes change often, breaking changes are expected, and there are no compatibility guarantees.

> [card] [Quickstart](https://docs.moonbounce.ai/quickstart.md)
>
> From a key to your first evaluation and feedback.

> [card] [How Moonbounce works](https://docs.moonbounce.ai/learn/how-it-works.md)
>
> The whole loop, with one running example, then each piece in turn.

> [card] [Connect your agent](https://docs.moonbounce.ai/connect-your-agent.md)
>
> Use the same operations as MCP tools.

> [card] [API reference](https://docs.moonbounce.ai/reference/overview.md)
>
> Every operation, generated from the OpenAPI document.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.