ManyCP MCP server

Submit an MCP server to many MCP directories and track each listing from an AI assistant

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

What data it sees

Do you need an account

No: the server works without sign-in

Submit an MCP server to many MCP directories and track each listing from an AI assistant

How to connect

How to connect ManyCP MCP server to Claude

The server address to paste into Claude settings:

https://manycp.com/mcp

Open Claude (claude.ai or the desktop app).
Go to Settings → Connectors.
Click Add custom connector.
Paste the server address copied below into Remote MCP server URL and click Add.
In a chat, click + → Connectors and switch the new connector on.

Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

Authorization

After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

How to connect ManyCP MCP server to ChatGPT

The server address to paste into ChatGPT settings:

https://manycp.com/mcp

Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
Open Settings → Connectors and click Create.
Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
In a new chat open + → Apps/Connectors and enable the connector.

OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

Authorization

On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

How to connect ManyCP MCP server to Cursor

The server address to paste into Cursor settings:

https://manycp.com/mcp

Fastest: click Open in Cursor below and confirm the prompt.

Manually:

In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
Save the file. The server appears in the MCP list; authorize it there if asked.

Authorization

If the service needs sign-in, an authorize button appears next to the server in the MCP list.

Server tool list (8)

Raw names from tools/list. Only developers need these.

list_directories	List every MCP directory ManyCP can file a server in (26 today). For each one you get: its id (pass ids to submit_server), name and URL, how ManyCP files it (Direct publish, Auto PR, Auto-indexed or Assisted), whether it fits local servers, remote servers or both, whether it must wait for another directory, and what ManyCP does step by step. Read-only, takes no input, works on every plan. Use it to answer 'where can I list my MCP server?' or to pick directory ids for submit_server.
list_servers	List the MCP servers on this ManyCP account. Each comes with its id and slug (every other tool takes either one as `server`), its repo and version, and a one-line status across the directories that fit it, for example '3 live, 1 PR open, 16 not started'. Read-only, works on every plan. Call this first when the user says 'my server' and you do not know which one they mean. Use get_server for the detail of one.
add_server	Add an MCP server to ManyCP from its public GitHub repo. ManyCP reads the repo (README, package.json or pyproject.toml, server.json, Dockerfile), finds npm, PyPI, Docker and remote endpoints, writes a draft listing (title, short description, categories, tags, tools, install config), saves it, and returns the listing, any warnings, and the plan for every directory that fits. It does not submit anywhere: review the listing, fix it with update_listing, then call submit_server. Adding a repo that is already added is an error that names the existing server. Needs a launch credit ($19 per server, once), Starter (3 servers) or Pro (unlimited).
get_server	Get one server's full listing and its status in every directory that fits it. Status is one of: live (ManyCP confirmed the listing exists), pr_open (a pull request is waiting on maintainers), submitted (sent, waiting), in_progress, queued, needs_update (the listing is behind the latest version), action_required (the user must do something, the reason is included), rejected, skipped, or not_started. Also returns links: pull requests, live listing pages, the dashboard page. Read-only, works on every plan. `server` accepts an id, a slug, a GitHub repo, or the exact name.
update_listing	Edit the listing ManyCP sends to every directory. Send only the fields you want to change; the rest stay as they are. Fields: title (max 100 characters), short_description (max 100 characters, because the official MCP Registry rejects longer), description (long text, Markdown is fine), categories (up to 5 ids), tags (up to 15, lowercase words), tools (the FULL list of {name, description}, it replaces the old list), homepage_url (an empty string clears it). Saves at once. Directories that already list the server keep their current text until the next sync or submit. Version, packages and repo come from GitHub and cannot be edited here. Needs a launch credit, Starter or Pro.
check_listings	Check which MCP directories already list a public GitHub repo. 10 of the 26 directories have a public lookup, so those come back listed or missing (with the listing URL when listed) and the rest come back unknown. No login and no ManyCP server needed, works on every plan and on repos you do not own. Results are cached for about 10 minutes. Use it before add_server to show the gap, or after submit_server to see what went live.
submit_server	File a saved server in MCP directories. Per directory the result says what happened. Direct publish (the official MCP Registry) publishes for real. Auto PR forks the target repo and opens a pull request from the user's GitHub account. Auto-indexed just waits for its source. Assisted directories get every field prepared plus a link, and a person finishes them (on Pro, ManyCP staff can). Leave `directories` out to send to every directory that fits and is ready, or pass ids from list_directories to pick some. ManyCP opens at most one pull request per repo per directory, so repeating the call is safe. A directory is only called live after ManyCP confirms it. The first submit of a server takes one launch credit ($19, bought once) unless a subscription covers it. Before the directories, ManyCP also opens ONE pull request on the user's own repo (never a fork) with the files directories look for: .mcp.json, glama.json, gemini-extension.json, server.json and the mcpName marker. cursor.directory, Glama and the Gemini CLI gallery wait until the user merges it, and the result says so. If the user has not allowed pull requests yet, the result is an error with a link they open once to grant GitHub access, then you call this again. Set consent_star to true only after the user agrees to let ManyCP star punkpeye/awesome-remote-mcp-servers on their GitHub account, because that list rejects pull requests from accounts that have not starred it.
sync_server	Re-read a server's GitHub repo right now: latest release or tag, package version, README, description, topics. If the version or listing changed, ManyCP updates the saved listing, marks directories that list an older version as needs_update, and starts update runs where a directory supports it (the official registry gets the new version; pull-request directories get an update PR only when the entry text changed). If nothing changed it says so. Paid plans also sync on their own (Pro every hour, Starter every day), so use this to sync now, for example right after a release. Needs a launched server or a paid plan.