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

# Quickstart

> Your first Datagate query in under 5 minutes

## 1. Get your API key

Sign up at [platform.getdatagate.com](https://platform.getdatagate.com) and generate an API key from the dashboard. Your key starts with `dg_live_`.

## 2. Install the SDK

<CodeGroup>
  ```bash Python theme={null}
  pip install datagate
  ```

  ```bash TypeScript theme={null}
  npm install datagate
  ```
</CodeGroup>

## 3. Discover datasets

<CodeGroup>
  ```python Python theme={null}
  from datagate import DatagateClient

  client = DatagateClient(api_key="dg_live_...")

  datasets = await client.list_datasets()
  for ds in datasets:
      print(f"{ds.name} — {ds.price_per_chunk} {ds.currency}/chunk")
      if ds.metadata_schema:
          for field in ds.metadata_schema:
              print(f"  Filter: {field['name']} ({field['type']})")
  ```

  ```typescript TypeScript theme={null}
  import { DatagateClient } from "datagate";

  const client = new DatagateClient({ apiKey: "dg_live_..." });

  const datasets = await client.listDatasets();
  for (const ds of datasets) {
    console.log(`${ds.name} — ${ds.price_per_chunk} ${ds.currency}/chunk`);
    if (ds.metadata_schema) {
      for (const field of ds.metadata_schema) {
        console.log(`  Filter: ${field.name} (${field.type})`);
      }
    }
  }
  ```
</CodeGroup>

## 4. Query

<CodeGroup>
  ```python Python theme={null}
  response = await client.query(
      text="machine learning transformers",
      dataset_ids=[ds.id for ds in datasets if ds.queryable],
      top_k=5,
  )

  for result in response.results:
      print(f"[{result.score:.4f}] {result.metadata}")

  await client.close()
  ```

  ```typescript TypeScript theme={null}
  const response = await client.query({
    text: "machine learning transformers",
    datasetIds: datasets.filter((ds) => ds.queryable).map((ds) => ds.id),
    topK: 5,
  });

  for (const result of response.results) {
    console.log(`[${result.score.toFixed(4)}]`, result.metadata);
  }
  ```
</CodeGroup>

## 5. Add filters (optional)

Use `metadata_schema` from step 3 to filter results per dataset:

<CodeGroup>
  ```python Python theme={null}
  response = await client.query(
      text="privacy regulations",
      dataset_ids=["ds-aaa", "ds-bbb"],
      filters={
          "ds-aaa": {"year": {"$gte": 2023}},
          "ds-bbb": {"category": {"$eq": "legal"}},
      },
  )
  ```

  ```typescript TypeScript theme={null}
  const response = await client.query({
    text: "privacy regulations",
    datasetIds: ["ds-aaa", "ds-bbb"],
    filters: {
      "ds-aaa": { year: { $gte: 2023 } },
      "ds-bbb": { category: { $eq: "legal" } },
    },
  });
  ```
</CodeGroup>

## Next steps

* [Authentication](/authentication) — API keys and JWT tokens
* [Metadata Filters](/guides/metadata-filters) — full filter syntax reference
* [MCP Integration](/mcp/overview) — connect Datagate to Claude Desktop