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

# EigenPal docs

> Build, test, and run AI workflows for repetitive document and data tasks.

EigenPal lets you define AI workflows as files, test them against datasets, and
run them from the dashboard, CLI, SDKs, or API. Most teams use it for document
processing: parse, extract, classify, transform, and route data. Workflows can
also call APIs, read websites, run scripts, and branch on logic.

The core workflow is:

<Steps>
  <Step title="Build">
    Define a [workflow](/concepts/workflows) as typed steps, or use a Git-backed
    [agent](/concepts/agents) for tasks that need code and per-input adaptation.
  </Step>

  <Step title="Evaluate">
    Pair it with a [dataset](/guides/build-a-dataset) and [evaluators](/concepts/evals) that encode
    the expected output.
  </Step>

  <Step title="Deploy">
    Start runs from the dashboard, CLI, SDKs, or API. Each run stores inputs, outputs, and traces.
    Experiments add evaluator scores.
  </Step>

  <Step title="Improve">
    Add failed real inputs to the dataset, fix the workflow, and compare the new experiment before
    shipping.
  </Step>
</Steps>

Workflows, agents, datasets, evaluators, and run artifacts are the main objects
you work with in EigenPal. See
[Eval-first development](/guides/eval-first-development) for a worked example.

## Start here

<CardGroup cols={2}>
  <Card title="Why EigenPal" icon="circle-question" href="/concepts/why">
    What it is for and when to use it.
  </Card>

  <Card title="How it works" icon="arrows-spin" href="/concepts/how-it-works">
    The build, evaluate, deploy, and improve flow.
  </Card>

  <Card title="Choose your path" icon="signs-post" href="/guides/choose-your-path">
    Pick CLI authoring, SDK integration, or raw REST API calls.
  </Card>

  <Card title="Your first workflow" icon="rocket" href="/guides/first-workflow">
    Scaffold, push, and run a workflow from the CLI in minutes.
  </Card>

  <Card title="Eval-first development" icon="flask-vial" href="/guides/eval-first-development">
    One example from initial workflow to regression test.
  </Card>
</CardGroup>

## Explore

<CardGroup cols={2}>
  <Card title="Design principles" icon="compass" href="/concepts/principles">
    How EigenPal handles tests, files, versions, and execution.
  </Card>

  <Card title="Workflow vs agent" icon="code-branch" href="/concepts/agents">
    When to use a fixed workflow and when to use a Git-backed agent.
  </Card>

  <Card title="Evaluations" icon="chart-line" href="/concepts/evals">
    Datasets, evaluators, and experiments.
  </Card>

  <Card title="Reviews" icon="clipboard-check" href="/concepts/reviews">
    Human verdicts, issue status, corrections, and monitoring.
  </Card>

  <Card title="Artifacts as files" icon="boxes-stacked" href="/concepts/artifacts">
    How workflows, datasets, evaluators, and agents round-trip.
  </Card>

  <Card title="Workflow steps" icon="diagram-project" href="/concepts/steps-overview">
    Every supported step type, with its config and output schema.
  </Card>

  <Card title="SDKs" icon="code" href="/sdks/overview">
    Trigger workflows and agents from TypeScript or Python.
  </Card>

  <Card title="REST API" icon="webhook" href="/api-reference">
    Raw HTTP routes, request bodies, response objects, and schemas.
  </Card>

  <Card title="CLI reference" icon="terminal" href="/cli/overview">
    The full `eigenpal` command surface.
  </Card>
</CardGroup>