SocialGrid — Instagram Scraper

Unofficial community Actor. Not affiliated with, endorsed by, or sponsored by Instagram or Meta.

Unified-schema Instagram profile data — flat, agent-ready JSON extracted via a real Playwright browser session, never a raw GraphQL response dump. Requires both a residential proxy and an authenticated Instagram session (see Input below) — Instagram rejects logged-out profile queries outright, regardless of IP or browser fingerprint.

Why this vs. a raw-blob scraper

Instagram's web_profile_info/GraphQL response is a deeply nested internal object (edge_followed_by, hd_profile_pic_url_info, dozens of feature flags) built for Instagram's own web client, not for downstream consumption. This actor extracts the same Author shape used across SocialGrid's YouTube/TikTok/Instagram scrapers, so an LLM agent or analytics pipeline can treat every platform identically. Fields Instagram doesn't expose (e.g. total_likes — Instagram has no such aggregate) come back as null, never 0.

What it does — and honestly, what it doesn't (yet)

Implemented and live-verified:

Mode	Description	Key input
profile	Account profile: display name, bio, avatar, follower/following/post counts, verified, private flag, external links	handle

Explicitly NOT implemented in this MVP — posts, comments, search, and trending are deferred, not attempted this pass. This actor is single-purpose (profile lookup) for now.

Why authentication is required (not just a proxy)

Modern Instagram no longer server-renders profile data into the page HTML at all — it's a client-side-only app shell. The real data lives behind an internal GraphQL/web_profile_info endpoint that returns "Unauthorized logged out query." for any request without a real, logged-in session — independent of proxy quality or browser fingerprint. There is no unauthenticated path to this data anymore.

This actor never automates Instagram's login form — that flow is far more bot-defended than viewing profiles and risks the account. Instead, you provide three cookies from your own real, manual browser login:

1. Log in to instagram.com normally, in a real browser.
2. Open DevTools → Application/Storage → Cookies → https://www.instagram.com.
3. Copy the values of sessionid, ds_user_id, and csrftoken.
4. Pass them as igSessionId, igDsUserId, igCsrfToken in the actor input (marked as secret fields — they're never logged or echoed).

These cookies belong to a real Instagram account, so treat them like any other credential: expect them to expire or be invalidated by Instagram at any time (password change, suspicious-activity flag, manual logout elsewhere), at which point the actor starts failing with the same "Unauthorized logged out query" error and the cookies need refreshing from a fresh manual login.

Example output (mode: "profile")

{
  "success": true,
  "platform": "instagram",
  "mode": "profile",
  "data": {
    "platform": "instagram",
    "id": "25025320",
    "handle": "instagram",
    "display_name": "Instagram",
    "bio": "Discover what's new on Instagram 🔎✨",
    "avatar_url": "https://scontent...cdninstagram.com/...",
    "verified": true,
    "followers": 685839159,
    "following": 256,
    "post_count": 8522,
    "total_likes": null,
    "external_links": [
      "https://www.youtube.com/watch?v=H6G9PcvHjj4",
      "https://open.spotify.com/episode/..."
    ],
    "is_private": false
  }
}

Input

• mode (required): profile — the only implemented mode
• handle (required for profile): account handle (instagram) or full profile URL
• proxyGroup (default RESIDENTIAL): Apify proxy group — Instagram requires a residential proxy to work at all
• igSessionId, igDsUserId, igCsrfToken (all required, secret): cookies from a real, human-completed login — see above

Usage

Apify CLI

apify actors call fulinlabs/socialgrid-instagram-scraper \
  --input '{"mode": "profile", "handle": "instagram", "igSessionId": "...", "igDsUserId": "...", "igCsrfToken": "..."}'

REST API

curl "https://api.apify.com/v2/acts/fulinlabs~socialgrid-instagram-scraper/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"mode": "profile", "handle": "instagram", "igSessionId": "...", "igDsUserId": "...", "igCsrfToken": "..."}'

Pricing

Benchmarked against comparable Store Instagram profile scrapers (Apify's official Instagram Profile Scraper at ~$1.60–2.60/1,000 profiles, third-party alternatives from $0.50–$20/1,000). Because this actor runs a full headless-Chromium session per call (heavier compute than a plain HTTP scraper) and depends on a maintained authenticated session, pricing sits at the higher end of that band:

• profile: $2.00 per 1,000 results

Reliability & known risks

• Standing operational dependency: the Instagram account behind the session cookies is real and subject to Meta's ban/challenge risk for automation-adjacent activity. Session expiry requires a human to manually re-log-in and refresh the cookies — this is not a one-time setup cost.
• Runs on Apify's official apify/actor-node-playwright-chrome:22 base image with a real Chromium browser — slower and more resource-intensive per call than a plain HTTP-based scraper.
• A 6-hour canary schedule re-runs the live test suite (getProfile against a known fixture + a nonexistent-handle error case) and alerts on failure.