Reviewer3 in your AI assistant

Connect Reviewer3 to Claude, ChatGPT, Cursor, or any MCP-compatible tool. Upload a manuscript, read the review, revise the draft, and format it for a journal — all without leaving your workflow.

1

Connect your AI assistant

Reviewer3's connection uses MCP (Model Context Protocol), an open standard for connecting AI assistants to outside tools securely. Connect your assistant below.

One click adds Reviewer3 to Cursor's MCP config for you.

Add to Cursor

Or add it manually

In Cursor, go to Settings → MCP and add a new server with this config:

{
  "mcpServers": {
    "reviewer3": {
      "url": "https://reviewer3.com/api/mcp"
    }
  }
}

What you can do once connected

  • Upload a manuscript
  • Find a past review by name
  • Check review status
  • Read the full review report
  • Check all references
  • Screen for fatal flaws
  • Edit the LaTeX manuscript
  • Format for a specific journal
  • Submit a revision

Troubleshooting

My AI assistant asks me to sign in but nothing happens

Make sure you've created a Reviewer3 account first (step 1 above), using the same email you authenticate with.

I don't see Reviewer3's tools after connecting

Some clients require you to explicitly enable or select the connector per conversation — check the + or tools menu near the message box.

Still stuck?

Reach out at support@reviewer3.com.

MCP Lets an Assistant Call Reviewer3 Directly

The Model Context Protocol is an open standard that lets an AI assistant call external tools directly. Instead of copying reports into a chat window and pasting responses back out, the assistant talks to Reviewer3 itself. It submits the file, waits for the pipeline to finish, and reads the structured report.

Authors, reviewers, and editors already work in general-purpose assistants. Asked to “review this paper,” one may produce useful feedback, but it has no way to check the underlying evidence or verify that every citation points to a real work.

Connecting your assistant to Reviewer3 gives it structured output from the verification pipeline. With that as context, it's better placed to work through revisions with you, answer questions, or improve a draft.

Compatible Clients

Any MCP-compatible client can connect today, including Claude, ChatGPT, and Cursor. Because MCP is an open standard rather than a per-vendor integration, a client that adds support later works with no change on our side.

API Access

Reviewer3 offers both. Use MCP when the work happens inside an AI assistant. Use the API when it happens inside a system, like submission software or an internal pipeline. Both run the same review and return the same report.