Analyze Instagram profiles from Claude or ChatGPT

The quick answer

Open Claude → Settings → Connectors → Add custom connector, paste https://www.instaseer.com/mcp, and sign in to InstaSeer when the consent screen appears. That is the entire setup. In ChatGPT, enable Developer mode under Apps & Connectors and create a connector with the same URL and OAuth. From that moment your assistant can answer questions like "how many followers does @nike have?" or "analyze @glossier's last 100 posts" with live public data instead of guesses.

Everything the assistant fetches is the same source-backed public data the InstaSeer web app shows: visible posts, dates, captions, public engagement fields, and posting calendars. Nothing requires an Instagram login, and nothing touches private accounts.

Set it up in your client

Claude (claude.ai and Desktop)

1. Settings → Connectors → Add custom connector.
2. Paste https://www.instaseer.com/mcp.
3. Click Connect, sign in to InstaSeer (a free account takes a minute), approve access.

ChatGPT

1. Settings → Apps & Connectors → enable Developer mode.
2. Create connector, MCP server URL https://www.instaseer.com/mcp, authentication OAuth.
3. Sign in to InstaSeer when prompted.

Claude Code, Cursor, VS Code

CLI and editor clients authenticate with an API key instead. Create one on the MCP install page — it fills ready-to-paste config for each client, including a one-click Cursor install link.

What your assistant can do once connected

• Free follower checks — instaseer_follower_count returns the public follower count, following, post count, verification badge, and engagement rate for any public handle. 100 fresh lookups per account per day, no credits.
• Free previews — instaseer_profile_preview returns headline profile fields before you commit credits to a deep report.
• Full timeline analysis — instaseer_profile_timeline pulls the visible public post timeline with per-post engagement and a posting calendar. About 1 credit per post, 50-credit minimum.
• Competitor comparisons — instaseer_compare_profiles puts up to five handles side by side.
• Balance checks — instaseer_account_usage reports your plan and remaining credits, free, any time.

Prompts that work well

• "How many followers does @nike have on Instagram right now?"
• "Analyze @glossier's last 100 Instagram posts. What's their posting cadence, and which formats get the most engagement?"
• "Compare @fentybeauty, @glossier, and @rarebeauty over their recent posts and summarize what each does differently."
• "Before you run anything paid, check how many InstaSeer credits I have left."

How billing works

MCP usage draws from the same credit balance as the InstaSeer web app — one pool, whichever surface you use. A free account includes 50 credits a month, which covers one full 50-post analysis plus unlimited free lookups and previews. Paid plans add capacity: Creator (1,500 credits, $9/mo), Plus (5,000, $29/mo), and Pro (18,000, $99/mo), with one-time credit packs that never expire for overflow.

When your assistant runs out of credits mid-research it does not dead-end: the tool result explains what happened and includes a secure Stripe checkout link, so you can top up and continue in the same conversation.

What this deliberately does not do

Every tool reads what Instagram shows logged-out visitors, nothing more. Private accounts stay private. Instagram publishes no follower lists to logged-out visitors, so no InstaSeer tool can judge which followers are real, and the engagement rate comes back as a plain number rather than an audience-quality score — across brands with genuine audiences it legitimately ranges from under 0.02% to over 7%, so it only means something between close peers. If an AI answer claims otherwise, it did not get that from InstaSeer.

FAQ