> Parel documentation. Index of every page: https://parel.si/docs/llms.txt. This page in HTML: https://parel.si/docs/tools/get_social_posts

# get_social_posts

> A public Instagram account's recent posts with captions, media and engagement counts, about 12 per page. ₹0.28 per page; a private or missing account is ₹0.

## What it does

The recent posts of a public Instagram account from its handle (without the @): for each post the caption, media type (photo, video, reel, carousel), like and comment counts, timestamp, media URLs and shortcode, about 12 per page, in `data.items[]`, with `next_max_id` for the next page. It cannot read private accounts, stories or comments; followers, bio and links are a separate profile tool that is not in the beta.

## Inputs

Generated from the contract captured from staging on 2026-11-05.

| Field | Type | Required | Description |
|---|---|---|---|
| `handle` | string | yes | Instagram handle. e.g. "natgeo" |
| `next_max_id` | string | no | Cursor to get next page of results. |

## What drives the cost

Pages. Each page is one call at the flat price; pass `next_max_id` from the response to fetch the next one.

## Price today

| | |
|---|---|
| Price today | **₹0.28** per call (`cost_paise: 28`) at ₹93.93 per US dollar on 17 November 2026 |
| Hold before the call | ₹0.42 (`hold_paise: 42`); when the call settles the real price is charged and the unused part is released |
| ₹0 when | it fails, or the provider answers but there is nothing in it (an empty list, no email, no page): `class: "ok_free"`, ₹0 |
| Timeout | 40 s, then `upstream_timeout` (₹0, retryable) |
| Over MCP | a headline tool: call `get_social_posts` directly |
| Over REST | `POST https://api.parel.si/v1/run` with `"tool": "get_social_posts"` |
| Confirmation | not needed (the estimate is under the ₹200 threshold) |

## Request

```bash
curl https://api.parel.si/v1/run \
  -H "Authorization: Bearer $PAREL_KEY" \
  -H "Content-Type: application/json" \
  -d '{"tool":"get_social_posts","input":{"handle":"natgeo"}}'
```

Over MCP, `get_social_posts` is a headline tool: Claude calls it directly with `handle` and, for the next page, `next_max_id`.

## What comes back

The response is large: 500 KB to 900 KB per page over REST. Over MCP it is cut to whole posts (valid JSON) with a `_truncated` note saying how many were kept; use the REST API when you need every post. A recorded answer from the source, cut to the top-level fields and two items, with most of each post's fields left out here:

```json
{
  "success": true,
  "next_max_id": "3947900…",
  "user": {
    "pk": "787132",
    "pk_id": "787132",
    "profile_grid_display_type": "default",
    "strong_id__": "787132",
    "id": "787132",
    "full_name": "Nationa…",
    "is_active_on_text_post_app": true,
    "is_private": false,
    "is_verified": true,
    "profile_pic_id": "3887353…",
    "profile_pic_url": "https:/…",
    "username": "natgeo"
  },
  "num_results": 12,
  "more_available": true,
  "items": [
    {
      "id": "3950509…",
      "pk": "3950509…",
      "image_versions2": {
        "candidates": [
          "… {6 field(s) truncated}",
          "… 2 more item(s) truncated"
        ]
      },
      "media_type": 1
    },
    "… 2 more item(s) truncated"
  ]
}
```

## Notes

- Big responses: 500 KB to 900 KB per page. Over MCP the list is cut to fit (valid JSON; _truncated says how many were kept); REST returns everything.
- About 12 posts per page; pass next_max_id from the response for the next page. Each page is a call.
- Private or missing accounts return no posts, which is free.

## Siblings

- [`get_reddit`](https://parel.si/docs/tools/get_reddit.md) for a community rather than an account.
- [`search_news`](https://parel.si/docs/tools/search_news.md) for press coverage of the same brand.
