Server data from the Official MCP Registry
Set up and manage conversion tracking by chatting with your AI assistant.
About
Set up and manage conversion tracking by chatting with your AI assistant.
Remote endpoints: streamable-http: https://app.converly.io/mcp
Security Report
Valid MCP server (1 strong, 1 medium validity signals). No known CVEs in dependencies. Imported from the Official MCP Registry.
Endpoint verified · Requires authentication · 1 issue found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
Permissions Required
This plugin requests these system permissions. Most are normal for its category.
How to Connect
Remote Plugin
No local installation needed. Your AI client connects to the remote endpoint directly.
Add this to your MCP configuration to connect:
{
"mcpServers": {
"io-converly-converly": {
"url": "https://app.converly.io/mcp"
}
}
}Documentation
View on GitHubFrom the project's GitHub README.
Converly MCP
Set up and manage your conversion tracking by chatting with your AI assistant.
Converly tracks form conversions and fires them to ad platforms (Google Ads, Meta, GA4, and more) when someone submits a form on your website. The Converly MCP server lets an AI assistant like Claude or ChatGPT do that work for you: connect ad platforms, build and publish conversion flows, install tracking, and check your conversions, all inside your Converly account.
This is a remote, hosted MCP server. There is nothing to install or run. You add it as a custom connector in your AI assistant and authorize it, and that is it.
Connector address: https://app.converly.io/mcp
Use with Claude Code and Cowork (plugin)
This repository is also a Claude Code plugin. It bundles the hosted connector together with two skills, one for guided conversion-tracking setup and one for diagnosing missing conversions.
/plugin marketplace add converlyio/converly-mcp
/plugin install converly@converly
The plugin connects to the same hosted server; you authorize with your Converly account on first use.
What you can do
Connect it, then just say what you want. For example:
- "Connect my Google Ads account and set up conversion tracking for my contact form."
- "Create a flow that sends a Lead event to Meta whenever someone submits my Typeform."
- "Did my form conversions fire to Google Ads yesterday? Show me what happened."
- "Why is my conversion not showing up in Google Ads?"
- "Publish my new flow and give me the snippet to install on my site."
- "Add our office IP so our own team visits do not get tracked."
Connect it
Claude (web or desktop)
- Open Settings, then Connectors, then Add custom connector.
- Paste the address:
https://app.converly.io/mcp - Authorize it. You log in to Converly (or create an account, which starts a free trial) and approve access.
- Start a new chat and try "Help me set up conversion tracking."
ChatGPT
- Enable Developer Mode under Settings, then Apps, then Advanced.
- Add a connector and paste the address:
https://app.converly.io/mcp - Authorize it, then start chatting.
Custom connectors require a supported plan on the AI side. Claude allows one custom connector on the Free plan and more on paid plans. ChatGPT requires a paid plan with Developer Mode.
How it stays safe
You stay in control the whole way:
- You authorize the connection through Converly's own login and a consent screen.
- The assistant cannot finish ad platform logins by itself. When you connect Google Ads or Meta, it hands you a secure link and you complete the login in your browser.
- Anything destructive or billing related requires explicit confirmation. Deleting a flow, disconnecting a platform, or changing your plan all require you to confirm by name first.
- Access is scoped, and every read or write is checked against your account on each request.
Tools
The server exposes 27 tools, grouped here by what they do:
Sites and install
list_sites, create_site, get_install_snippet, get_install_status
Flows (a flow fires a conversion when a form is submitted)
list_flows, get_flow, create_flow, update_flow, validate_flow, publish_flow, unpublish_flow, delete_flow
Destinations (your ad platforms)
list_destinations, list_destination_types, connect_destination, check_handoff_status, disconnect_destination, list_conversions
Conversions and events
list_events, get_event_detail, send_test_event
Discovery
list_trigger_types, list_action_types
Account and billing
get_subscription, list_plans, change_plan
Internal traffic
add_internal_traffic_rule
The server also ships guided prompts (one-click workflows in clients that support them): set up conversion tracking, connect an ad platform, and check recent conversions.
Links
- Converly for AI assistants: https://converly.io/mcp
- Converly home: https://converly.io
- Pricing and free trial: https://converly.io/pricing
Security and data handling
Declared for reviewers and anyone auditing this plugin before installing it.
This plugin contains no executable code. It is a set of manifests plus two Markdown skill files. There are no install or lifecycle hooks, no bundled binaries or minified code, no postinstall steps, and nothing is downloaded or run at install time.
Network endpoints contacted:
https://app.converly.io/mcp— the hosted Converly MCP server, and the only endpoint this plugin causes an agent to call. It is declared in.mcp.jsonas a remote streamable HTTP server.
Credentials: none are stored in, read by, or bundled with this plugin. Access is authorized through OAuth against the user's own Converly account on first use, and the resulting token is held by the MCP client, not by anything in this repository. The plugin never reads local files, environment variables, SSH keys, or .env files.
Ad platform credentials never pass through this plugin or the MCP server. Connecting a destination such as Google Ads or Meta returns a Converly-hosted link that the user opens in their own browser to authorize the platform directly.
Scope: the MCP server exposes named, purpose-scoped tools. There is no shell-execution tool and no catch-all passthrough. Destructive operations (deleting a flow, disconnecting a platform, changing the plan) require explicit user confirmation, and tools carry read-only and destructive annotations so clients can prompt appropriately.
Telemetry: none beyond the API requests the user's own actions generate.
About this repository
This repository is the public home and documentation for the Converly MCP integration. It does not contain the server code. The server is a hosted part of Converly and runs at https://app.converly.io/mcp. For questions or issues with the MCP, open an issue here or contact Converly support.
Reviews
No reviews yet
Be the first to review this server!
More Developer Tools MCP Servers
Git
Freeby Modelcontextprotocol · Developer Tools
Read, search, and manipulate Git repositories programmatically
Fetch
Freeby Modelcontextprotocol · Developer Tools
Web content fetching and conversion for efficient LLM usage
Toleno
Freeby Toleno · Developer Tools
Toleno Network MCP Server — Manage your Toleno mining account with Claude AI using natural language.
mcp-creator-python
Freeby mcp-marketplace · Developer Tools
Create, build, and publish Python MCP servers to PyPI — conversationally.
MCP Marketplace
Freeby mcp-marketplace · Developer Tools
Search and install MCP servers from inside your AI client.
MarkItDown
Freeby Microsoft · Content & Media
Convert files (PDF, Word, Excel, images, audio) to Markdown for LLM consumption