Google Tag Manager MCP Server
Enables AI agents to manage Google Tag Manager accounts, containers, workspaces, tags, triggers, variables, and versions via the Tag Manager API v2.
README
Google Tag Manager MCP Server
Built by Growthers.io — Data-driven growth agency.
An open-source MCP (Model Context Protocol) server that gives AI agents full access to the Google Tag Manager API v2.
Works with Claude Code, Claude Desktop, Cursor, and any MCP-compatible client. Manage accounts, containers, workspaces, tags, triggers, variables, consent mode, and container versions — all from your AI assistant.
No external services required. Runs locally on your machine using your own Google credentials.
Prerequisites
- Python 3.10+
- pipx (recommended) or pip
- A Google Cloud project with the Tag Manager API enabled
- gcloud CLI for credential setup
Setup
1. Enable the Tag Manager API
- Go to Google Cloud Console
- Select or create a project
- Go to APIs & Services > Enable APIs
- Search for Tag Manager API and enable it
2. Authenticate
gcloud auth application-default login \
--scopes="https://www.googleapis.com/auth/cloud-platform,https://www.googleapis.com/auth/tagmanager.readonly,https://www.googleapis.com/auth/tagmanager.edit.containers,https://www.googleapis.com/auth/tagmanager.edit.containerversions"
This creates a credentials file at ~/.config/gcloud/application_default_credentials.json.
Tip: If the default gcloud client blocks some scopes, create a custom OAuth Client ID in your Google Cloud project (APIs & Services > Credentials > Create OAuth Client > Desktop App), then use
--client-id-file=<path-to-client-secret.json>with the command above.
3. Install
Option A: With pipx (recommended)
pipx install git+https://github.com/CarC96/google-tag-manager-mcp.git
Option B: From source
git clone https://github.com/CarC96/google-tag-manager-mcp.git
cd google-tag-manager-mcp
pip install .
4. Configure your MCP client
Add to your MCP client configuration (e.g. ~/.claude.json for Claude Code):
{
"mcpServers": {
"gtm-mcp": {
"type": "stdio",
"command": "pipx",
"args": ["run", "gtm-mcp"],
"env": {
"GOOGLE_APPLICATION_CREDENTIALS": "<path-to>/application_default_credentials.json"
}
}
}
}
Replace <path-to> with the actual path to your credentials file (typically ~/.config/gcloud/application_default_credentials.json).
How it works
Despite the name "server", this runs as a local process on your machine. Your MCP client launches it as a subprocess and communicates via stdin/stdout. No network ports are opened, no cloud hosting is needed — it's essentially a plugin.
MCP Client (Claude Code, etc.) ←— stdin/stdout —→ gtm-mcp (local process) ——→ Google Tag Manager API
Available Tools (30)
Accounts
| Tool | Description |
|---|---|
list_accounts |
List all GTM accounts |
get_account |
Get account details |
list_user_permissions |
List permissions for an account |
Containers
| Tool | Description |
|---|---|
list_containers |
List all containers in an account |
get_container |
Get container details |
get_container_snippet |
Get the install snippet for a website |
Workspaces
| Tool | Description |
|---|---|
list_workspaces |
List workspaces in a container |
get_workspace |
Get workspace details |
get_workspace_status |
Check for conflicts and modified entities |
create_workspace |
Create a new workspace |
Tags
| Tool | Description |
|---|---|
list_tags |
List all tags |
get_tag |
Get tag details |
create_tag |
Create a new tag |
update_tag |
Update a tag |
delete_tag |
Delete a tag |
Triggers
| Tool | Description |
|---|---|
list_triggers |
List all triggers |
get_trigger |
Get trigger details |
create_trigger |
Create a new trigger |
update_trigger |
Update a trigger |
delete_trigger |
Delete a trigger |
Variables
| Tool | Description |
|---|---|
list_variables |
List all variables |
get_variable |
Get variable details |
create_variable |
Create a new variable |
update_variable |
Update a variable |
delete_variable |
Delete a variable |
Built-in Variables
| Tool | Description |
|---|---|
list_built_in_variables |
List all enabled built-in variables |
Versions
| Tool | Description |
|---|---|
list_version_headers |
List all version headers (summary) |
get_live_version |
Get the published version with all entities (great for audits) |
get_version |
Get a specific version by ID |
Use Cases
- Consent Mode Audits — Analyze all tags and their consent settings in seconds
- Bulk Tag Management — Create, update, or delete tags/triggers/variables programmatically
- GDPR Compliance — Verify consent configurations across all tags
- Container Migrations — Read and replicate configurations across containers
- Documentation — Auto-generate reports of your GTM setup
About Growthers.io
Growthers.io is a data-driven growth agency specializing in analytics, tracking infrastructure, and marketing automation. We build tools that make digital marketing more efficient and compliant.
License
Apache-2.0
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。