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

# Welcome

> Natural-language image search for agents — every answer carries a machine-readable rights answer.

Lightdrift is an image search engine built to be called by software, not browsed by people. Describe the image you need — in text, with a reference image, or both — and get ranked, usable images back. Every result carries a machine-readable rights answer, so your agent knows what it can do with the file before it uses it.

Retrieval only: no generation, no checkout, no marketplace. One API, billed per query.

## What it does

* **Text search** — natural-language intent, including design direction: "dark and moody, room for a headline on the left."
* **Search by image** — upload a reference, get visually and semantically similar assets.
* **Multimodal search** — a reference image plus text that steers it: "like this but at night."
* **Negative constraints** — say what to exclude: "no people", "no text."
* **Rights on every result** — license, permissions, attribution string, and delivery mode, as declared by the source.

## Start here

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Get an API key and run your first search in under a minute.
  </Card>

  <Card title="API reference" icon="code" href="/api-reference/introduction">
    Every endpoint, with an interactive playground and code samples.
  </Card>

  <Card title="Rights answers" icon="scale-balanced" href="/guides/rights">
    How to read the rights object and what Lightdrift does and does not promise.
  </Card>

  <Card title="Use with agents" icon="robot" href="/guides/agents">
    Connect Lightdrift to Claude, Cursor, or your own agent via MCP.
  </Card>
</CardGroup>