# arXiv · Harvest records

Get developer and code data with arXiv · Harvest records from Alexandria through one Spiral API key. Bulk or incremental synchronisation of arXiv metadata by date window, or the only surface that states a paper's license. For topical search use `search_papers`. It runs in the background and returns structured JSON. Pay as you go, with no subscription.

- Provider: Alexandria
- Category: Developer
- Runs: In the background
- Returns: json
- Tool id: `alexandria` / `arxiv-org/papers/harvest_records`
- Page: https://spiral.ai/tools/alexandria/arxiv-org/papers/harvest_records

## Pricing

$0.00873 per call.

## Limits

- Longest queue wait: 1 hour
- Run deadline: 5 minutes
- Longest run, end to end: 225 minutes
- Largest inline input: 1 MB
- Largest input file: 2 GB
- Largest output: 5 GB

## Input

| Field | Kind | Allowed | Default | Description |
| --- | --- | --- | --- | --- |
| `set` | text |  | — | OAI set spec from `list_sets`, e.g. `cs`, `econ`, `physics:hep-th`. Omit for all of arXiv. Pattern: ^[a-zA-Z][a-zA-Z0-9:.-]*$. |
| `from` | text |  | — | Earliest datestamp (`YYYY-MM-DD`, inclusive). Pattern: ^[0-9]{4}-[0-9]{2}-[0-9]{2}$. |
| `limit` | number | 1 to 1000 | — | Records to return in this call (1-1000; default 100). |
| `until` | text |  | — | Latest datestamp (`YYYY-MM-DD`, inclusive). Pattern: ^[0-9]{4}-[0-9]{2}-[0-9]{2}$. |
| `cursor` | text |  | — | `next_cursor` from a previous call. When given, `set`, `from` and `until` must be omitted. |

## Example

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "alexandria",
    "model": "arxiv-org/papers/harvest_records",
    "input": {
      "set": "<set>"
    }
  }'
```

Try it in the Playground: https://spiral.ai/start?provider=alexandria&model=arxiv-org%2Fpapers%2Fharvest_records

## Common questions

### arXiv · Harvest records

**How much does arXiv · Harvest records cost?**

$0.00873 per call. You pay as you go from one Spiral balance, with no subscription, and a failed run is not charged.

**How do I use arXiv · Harvest records through an API?**

Send `POST /v1/runs` with `"provider": "alexandria"`, `"model": "arxiv-org/papers/harvest_records"` and the input fields on this page, using one Spiral API key. It runs in the background: poll the run, or name a webhook to be told when it finishes.

**Do I need an account with Alexandria to use arXiv · Harvest records?**

No. Spiral runs it on its own Alexandria account and charges your Spiral balance.

**Can my AI agent use arXiv · Harvest records?**

Yes. Agents find and run it through Spiral's MCP server, CLI or agent skill, with the same API key.

