Instagram MCP server for AI agents

What you can do with Instagram MCP

Describe the task in your own words — the agent pulls the media, the profile that posted it and the audience around it

• Instagram profile

  Bio, follower and post counts, plus verified and business badges.

• Account search

  Look up an account by name when you do not remember the exact username.

• Posts and reels

  The latest posts and reels of any public account, with media and captions.

• Post by URL

  Paste a link to a post, reel or carousel and get every slide, the caption and the author.

• Comments

  Comments under a post or reel, each with its author and date.

• Likers

  Who liked a post or reel, with usernames and profile pictures.

• Followers

  The follower list of an account, loaded page by page.

• Following

  The accounts a profile follows, loaded page by page.

• Active stories

  Every story currently live on an account, saved before it expires.

• Story by URL

  One story from a link: the media, the author and when it expires.

• Highlights

  Highlight collections and every story saved in them.

• Download videos

  Direct download links for videos from posts, reels and IGTV, watermark-free.

Full Instagram access through a single MCP

Reels, stories, highlights, photos, videos and carousels — the same tools are available in a chat app, a code editor or your own scripts.

{
  "mcpServers": {
    "picuki-mcp": {
      "type": "http",
      "url": "https://api-wh.picuki.site/instagram/mcp"
    }
  }
}

Connect Instagram MCP to your AI tool

Choose your client and follow the steps below

Universal setup Universal ChatGPT ChatGPT Claude Claude

🔌 Suitable for any MCP-compatible client · only the location of the config file changes.

1. In your client, open the MCP settings and locate the config file — most often mcp.json.
2. Put Picuki into the mcpServers section:
   mcp.json Copy

   {
     "mcpServers": {
       "picuki-mcp": {
         "type": "http",
         "url": "https://api-wh.picuki.site/instagram/mcp"
       }
     }
   }

3. Save the changes, then restart the client to load the server.
4. Try a request in the chat:
   • Get the video from this post: instagram.com/p/…
   • Download all current stories from this account

One key with all features included

• Pro

  $50.00 /mo

  What you get:

  • 50,000 requests per month
  • Fixed limit — no unexpected charges
  • Proxies included
  • MCP access
  Try for free
• Most popular

  Ultra

  $150.00 /mo

  What you get:

  • 250,000 requests per month
  • Fixed limit — no unexpected charges
  • Proxies included
  • MCP access
  Try for free

• Mega

  $250.00 /mo

  What you get:

  • 1,000,000 requests per month
  • Fixed limit — no unexpected charges
  • Proxies included
  • MCP access
  Try for free

To subscribe, ask for a payment link right in the chat where Picuki is connected — ChatGPT, Claude or any other client — and the assistant will send it to you.

Need higher limits? Get in touch for a custom plan.

Frequently asked questions

What to know before connecting Picuki to your AI agent.

• What is MCP?

  MCP (Model Context Protocol) is an open standard that lets AI assistants use external tools and data sources. The server provides the tools, and the assistant chooses when to use them.

• Which clients are supported?

  Any MCP-compatible client. The setup guide above walks through ChatGPT and Claude — web, desktop and Claude Code — and gives a universal config for all other clients.

• Can I access private accounts?

  No. Just like the Picuki website, the MCP works with public content only.

• How does MCP differ from the REST API?

  The data is the same, the consumer is not. The REST API is meant for code you write yourself, while MCP lets an AI agent decide on its own when to fetch or download something.