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

# TypeScript SDK

> Define evals, mock dependencies and validate results in TypeScript

`anpord` is the TypeScript package for defining and running coding-agent evals. Cases, mocks and validators live in your repository alongside the code they test.

```sh theme={null}
npm install anpord
```

Start with the [quickstart](/getting-started) to define a suite and run it on Anpord.

## Package exports

| Import              | Use                                                                 |
| ------------------- | ------------------------------------------------------------------- |
| `anpord`            | [`defineEval`, workspace sources and validator types](/evals/cases) |
| `anpord/mcp`        | [`server`, `tool`, `resource`](/evals/mocks#mcp)                    |
| `anpord/cli`        | [`cli`, `command`](/evals/mocks#cli)                                |
| `anpord/api`        | [`api`, `endpoint`, `withApi`](/evals/api-mocks)                    |
| `anpord/validators` | [`judge`](/evals/judges)                                            |
| `anpord/eval`       | [`compileEval`](/guides/run-from-code)                              |
| `anpord`            | [`Anpord` client for runs, results and prompts](/sdk/client)        |

Definitions take objects. Handlers use ordinary TypeScript functions and promises. Standard Schema infers mock inputs and checks values at runtime; examples use Zod. You do not need Effect imports or a separate model client.

## One definition, multiple ways to run

Keep suites in `*.eval.ts` files. The CLI compiles them locally, including imported fixtures and validators, then submits them to Anpord's hosted infrastructure:

```sh theme={null}
npx anpord eval ./smoke.eval.ts
```

The [GitHub Action](/guides/ci) runs the same command. To submit from application code, use [`compileEval` with the API client](/guides/run-from-code). Results retain the calls, validator evidence and captured source for that run.

## Connections

Configure `ANPORD_API_KEY` for calls to Anpord. Save model credentials in [Settings > Harnesses](/connections); they do not belong in eval definitions. Hosted sandboxes and mocked product services need no separate keys by default.

For agent setup, start with [Claude Code](/guides/claude-code) or the [supported agents](/evals/variants).