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

# Captar

> Runtime control for OpenAI apps, with traces, budgets, and manual review built into the docs path.

<div
  style={{
position: "relative",
overflow: "hidden",
borderRadius: "28px",
border: "1px solid color-mix(in srgb, currentColor 14%, transparent)",
background: "color-mix(in srgb, currentColor 2%, transparent)",
boxShadow: "0 24px 60px rgba(0, 0, 0, 0.12)",
padding: "clamp(1.5rem, 4vw, 2.5rem)",
}}
>
  <div style={{ display: "flex", alignItems: "flex-start", justifyContent: "space-between", gap: "1rem", flexWrap: "wrap" }}>
    <div style={{ display: "flex", alignItems: "center", gap: "0.9rem" }}>
      <img
        src="https://mintcdn.com/captar/zWGZ6B1tNvZMbkx9/logo.png?fit=max&auto=format&n=zWGZ6B1tNvZMbkx9&q=85&s=cc932b30f07ad04dcd7b642d52be583d"
        alt="Captar logo"
        width="58"
        height="58"
        style={{
      borderRadius: "1rem",
      boxShadow: "0 12px 28px rgba(0, 0, 0, 0.15)",
    }}
        data-path="logo.png"
      />

      <div style={{ display: "flex", flexDirection: "column", gap: "0.2rem" }}>
        <span
          style={{
        fontSize: "1.35rem",
        lineHeight: 1,
        letterSpacing: "-0.04em",
        color: "var(--mintlify-content-primary)",
        fontFamily: "serif",
        fontWeight: 700,
        fontStyle: "italic",
      }}
        >
          Captar
        </span>

        <span
          style={{
        textTransform: "uppercase",
        letterSpacing: "0.18em",
        fontSize: "0.72rem",
        color: "var(--mintlify-muted)",
      }}
        >
          Runtime control notebook
        </span>

        <span style={{ fontSize: "1rem", color: "var(--mintlify-content-secondary)" }}>
          SDK-first docs for traces, budgets, tools, and review.
        </span>
      </div>
    </div>

    <div
      style={{
    padding: "0.6rem 0.9rem",
    borderRadius: "999px",
    border: "1px solid color-mix(in srgb, currentColor 12%, transparent)",
    fontSize: "0.85rem",
    color: "var(--mintlify-content-secondary)",
    background: "color-mix(in srgb, currentColor 1.5%, transparent)",
  }}
    >
      OpenAI and OpenAI-compatible only
    </div>
  </div>

  <div style={{ maxWidth: "52rem", marginTop: "2.2rem" }}>
    <p style={{ margin: 0, color: "var(--mintlify-content-primary)", fontWeight: 600, letterSpacing: "0.16em", textTransform: "uppercase", fontSize: "0.78rem" }}>
      V1 docs
    </p>

    <h1
      style={{
    margin: "0.7rem 0 0",
    fontSize: "clamp(2.6rem, 6vw, 4.8rem)",
    lineHeight: 0.98,
    letterSpacing: "-0.05em",
  }}
    >
      Captar gives OpenAI apps a runtime control layer that feels calm, precise, and easy to inspect.
    </h1>

    <p
      style={{
    margin: "1rem 0 0",
    maxWidth: "46rem",
    color: "var(--mintlify-content-secondary)",
    fontSize: "1.02rem",
    lineHeight: 1.85,
  }}
    >
      Use the TypeScript SDK to reserve budget before a request, wrap your existing client instead of replacing it,
      track tools inside the same trace, and promote the best runs into datasets and manual eval review.
    </p>
  </div>

  <div style={{ marginTop: "1.5rem" }}>
    <Callout title="What ships in v1" type="info">
      Captar v1 focuses on traces, datasets, and manual review. The docs intentionally stay narrow so the SDK path is
      easy to follow and the platform story stays grounded in what is already available.
    </Callout>
  </div>
</div>

<Cards>
  <Card title="Start the quickstart" href="/getting-started/quickstart">
    Create a session, wrap OpenAI, and record a tool call in one flow.
  </Card>

  <Card title="Learn the runtime model" href="/core-concepts/sessions-and-budgets">
    See how budgets, policy, trace context, and usage reconciliation fit together.
  </Card>

  <Card title="See the operational loop" href="/platform/traces">
    Inspect traces, export strong runs, and review them manually.
  </Card>
</Cards>

## What Captar helps with

<Steps>
  <Step>Keep spend local to the request instead of depending on a proxy or gateway.</Step>
  <Step>Apply call and tool guardrails before work reaches a provider or external system.</Step>
  <Step>Keep the span tree rich enough that trace review is useful later.</Step>
  <Step>Move the best examples into datasets so the review loop stays repeatable.</Step>
</Steps>

## Reading order

* [Overview](/getting-started/overview)
* [Installation](/getting-started/installation)
* [Quickstart](/getting-started/quickstart)
* [OpenAI wrapping](/core-concepts/openai-wrapping)
* [Manual evals](/platform/manual-evals)