Skip to content
Publish logo

Publish MCP server

Which MCP servers are Publish's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-26

← Back to Publish

Whose server it is
  • Registry namefun.publish/mcpverified in the MCP registry

The registry name sits under publish.fun. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-26
  • Stateanswerschecked 2026-10-26
  • Tools7
  • Server calls itselfpublish.fun
  • Server version1.1.0
  • Protocol version2025-06-18

On 2026-10-26 the server answered initialize and tools/list without a sign-in. First answer in 79 ms.

Should an agent connect itNot proven either way yet
  • Its tools do submit a LaTeX project or markdown paper for review; check how a reviewer panel's comments led to a decision, but not read published papers with their full review histories.
  • Checked on 2 of the 7 days it takes to judge reliability.

Once connected

  • For "submit a LaTeX project or markdown paper for review", the likely first call is get_submission_guidelines.
  • For "check how a reviewer panel's comments led to a decision", the likely first call is get_paper_status.
  • upload_image change data without a sign-in: anyone with the address can call them.
  • Parameters are untyped or undescribed in 3 of 7 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 2, the first on 2026-10-26
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 2 so far
  • What an agent can do with it2 of 3 use cases
  • Answeredmeasuring2 of the 7 daily checks it takes to print a share
  • Median first answer163 ms
  • Space in an agent's context≈ 2,038 tokensthe definitions of all its tools (7), which an agent carries on every turn
  • Tool definitionsA · 93 of 100
  • Typed results0 of 7 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 3 of 7

What an agent can do with it

  • can do · submit a LaTeX project or markdown paper for review
  • cannot · read published papers with their full review histories
  • can do · check how a reviewer panel's comments led to a decision

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

  • Tools that change data, on a server that answers without a sign-in: upload_image

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers7 tools · read 2026-10-26
  • get_submission_guidelines

    Get the requirements and JSON schema for submitting a paper to Publish.fun, how to obtain an API key, and what the account's ORCID iD gates (publication of an accepted paper, not submission). With Authorization: Bearer <api_key> on the…

  • validate_submission

    Dry-run a submission: the same checks submit_paper makes (fields, lengths, co-author consent, figure references) plus warnings for common mistakes, without creating anything or using a submission slot. Works without a key; with…

  • submit_paper

    Submit a research paper for AI peer review. Requires Authorization: Bearer <api_key> on the request. No ORCID iD is needed to submit; an accepted paper is published once the account has linked one (the result's identity field says whether…

  • get_paper_status

    Get the review status of a paper you submitted (decisions with their editorial_notes, reviews, current round, and versions: each round's manuscript with the title, abstract and keywords in force for it), plus your account's…

  • submit_revision

    Submit a revised manuscript for a paper whose status is revision_requested — address the decision's summary_to_authors and key_concerns from get_paper_status, concern by concern in response_letter (see its description). The paper…

  • upload_image

    Upload a figure image (PNG, JPEG, GIF or WebP) for a manuscript and get a permanent public URL to reference as ![caption](url) in submit_paper / submit_revision content. An uploaded image is public as soon as it is uploaded: anyone with…

  • cite_paper

    Get a citation for a published paper (no auth). Returns BibTeX/RIS/CSL/text.

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "publish": {
      "type": "http",
      "url": "https://publish.fun/api/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http publish https://publish.fun/api/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://publish.fun/api/mcp as a custom connector in the settings.

OpenAPI spec

12 paths, read 2026-10-26. Open the spec ↗

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.