Chat-first form service with 127 MCP tools. Create, manage, and analyze forms entirely through conversation.

formlova-mcp

Connect FORMLOVA to any MCP-compatible AI client without hand-writing setup files.

FORMLOVA is a chat-first form operations service. Most form tools stop at creation — FORMLOVA starts after publish. It covers the full post-publish lifecycle: response management, follow-up emails, reminders, A/B tests, email sequences, LINE messaging, analytics, team management, and more — all from your AI chat.

138 MCP tools across 24 categories. Works with Claude, ChatGPT, Codex, Claude Code, Cursor, Gemini CLI, Windsurf, and any MCP-compatible client.

formlova-mcp is not the FORMLOVA server. It is a small CLI helper for setup snippets, project-level config files, and a quick connectivity check.

The fastest path

Most users do not need this package.

If your AI client already has an MCP add screen, use this URL directly:

https://formlova.com/api/mcp

Reviewing setup instructions and registering the URL does not require a FORMLOVA account. On first real use, you can log in or sign up during authentication.

If you want a CLI helper instead, start here:

npx -y formlova-mcp print --client all

If you want project-level config files written for supported developer tools:

npx -y formlova-mcp install --client all --scope project --yes

This writes config files only. Authentication still begins inside the client on first use, and some clients require a restart or reconnect before the tools appear.

What this package is for

• Print copy-paste setup snippets for supported clients
• Generate project-level config files for common developer tools
• Check whether the FORMLOVA MCP URL is reachable from your machine
• Standardize local MCP setup across a team or repo

Supported clients

This package provides setup helpers for:

• ChatGPT
• Claude
• Codex
• Claude Code
• Cursor
• Gemini CLI
• Windsurf

Any MCP-compatible client can connect with the same URL — the list above is not a limitation:

https://formlova.com/api/mcp

Command overview

Command	Purpose	Writes files
print	Show setup snippets for one or more clients	No
install	Write supported project-level config files	Yes
doctor	Check MCP URL reachability and local config presence	No

print

Print setup snippets only. No files are written.

npx -y formlova-mcp print --client codex
npx -y formlova-mcp print --client claude-code
npx -y formlova-mcp print --client cursor
npx -y formlova-mcp print --client chatgpt
npx -y formlova-mcp print --client gemini-cli

install

Write project-level setup files for supported developer tools.

npx -y formlova-mcp install --client all --scope project --yes

install does not complete OAuth or client-side authentication for you.

Current write targets:

Client	Path written by install
Codex	.codex/config.toml
Claude Code	.mcp.json
Cursor	.cursor/mcp.json

For ChatGPT and Gemini CLI, install prints setup guidance because there is no stable local config file target in this package.

doctor

Check MCP URL reachability and common local config files.

npx -y formlova-mcp doctor

Client setup references

ChatGPT

• Enable developer mode first
• Then open Settings > Apps > Advanced settings (Create app) and add the MCP server URL
• Docs: OpenAI remote MCP guide

Claude

• Open Settings > Connectors
• Add the remote MCP server URL
• Docs: Claude remote MCP guide

Codex

codex mcp add formlova --url https://formlova.com/api/mcp

Claude Code

claude mcp add --transport http formlova https://formlova.com/api/mcp

Cursor

You can use the normal MCP UI, or a config file such as:

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

Gemini CLI

gemini mcp add --transport http formlova https://formlova.com/api/mcp

What it does not do

• It does not run a local FORMLOVA MCP server
• It does not replace the normal web-based setup flow for ChatGPT or Claude
• It does not complete client-side authentication or OAuth on your behalf
• It does not auto-write user-level config yet

Requirements

• Node.js 18.17+

Notes

Client-side setup menus and steps may change without notice. When in doubt, ask your AI client how to add an MCP server, then use:

https://formlova.com/api/mcp

Links

• Product: formlova.com
• Setup guide: formlova.com/en/setup
• Workflow Place: formlova.com/en/workflows
• Blog: formlova.com/en/blog
• Public repo: github.com/formlova/formlova-mcp
• Product Hunt: producthunt.com/products/formlova