mcp-search-console
Multi-account Google Search Console MCP server enabling AI assistants to query multiple GSC accounts by name in a single session, with support for OAuth and service accounts.
README
mcp-search-console
Multi-account Google Search Console MCP server. Connect any number of GSC accounts to Claude, Cursor, Codex, or any MCP-compatible AI assistant — and query them by name in the same session.
# Install: uvx mcp-search-console-multi
# Ask your AI:
"Show me the top queries for my-site last month"
"Compare client-acme's performance between Q1 and Q2"
"Check indexing issues on client-beta's 5 product pages"
Why this one?
Most GSC MCP servers support one account per server process. This one lets you configure multiple accounts (your own sites + client sites) and switch between them per tool call — no restart needed.
| Feature | This server | Others |
|---|---|---|
| Multiple accounts | Yes — named, switchable | No — one per process |
| OAuth + service account | Both, mixed per account | Usually one type |
| Auto token refresh | Yes | Sometimes |
| Rate limit retry | Yes — exponential backoff | No |
| Destructive op guard | Yes — env flag required | Sometimes |
| SSE transport (remote) | Yes | Varies |
Quickstart (uvx — no clone needed)
1. Create your accounts config:
mkdir -p ~/.config/mcp-search-console
cp accounts.example.json ~/.config/mcp-search-console/accounts.json
# Edit it — add your accounts
2. Add to your MCP client config:
{
"mcpServers": {
"search-console": {
"command": "uvx",
"args": ["mcp-search-console-multi"],
"env": {
"GSC_ACCOUNTS_CONFIG": "/Users/you/.config/mcp-search-console/accounts.json"
}
}
}
}
3. Restart your AI client. Done.
Accounts config
Copy accounts.example.json and edit it:
{
"default": "my-site",
"accounts": {
"my-site": {
"type": "oauth",
"client_secrets_file": "~/.config/mcp-search-console/client_secrets.json",
"token_file": "~/.config/mcp-search-console/my-site.token"
},
"client-acme": {
"type": "service_account",
"credentials_file": "~/.config/mcp-search-console/acme.json"
}
}
}
Set GSC_ACCOUNTS_CONFIG to its path, or put it at ~/.config/mcp-search-console/accounts.json (default).
OAuth setup
- Google Cloud Console → create project
- Enable the Search Console API
- Credentials → Create → OAuth client ID → Desktop app
- Download as
client_secrets.json - On first use, a browser window opens for you to authorise — token is saved automatically
Service account setup
- Google Cloud Console → Credentials → Create → Service Account
- Keys tab → Add Key → JSON → download
- In GSC, add the service account email as a user on each property
Using multiple accounts
Every tool accepts an optional account parameter. Omit it to use your default.
"Show top queries for my-site" # uses default
"Show top queries for client-acme" # uses named account
"Compare client-beta performance Jan vs Feb" # named account
Or set the default mid-session:
"Switch to client-acme as my default account"
Available tools
Account management
| Tool | What it does |
|---|---|
list_accounts |
Show all configured accounts and which is default |
set_default_account |
Change the default account |
reauthenticate |
Re-run OAuth flow or reload credentials for an account |
Properties
| Tool | What it does |
|---|---|
list_properties |
List all GSC properties |
get_site_details |
Verification + permission details for a property |
Search analytics
| Tool | What it does |
|---|---|
get_search_analytics |
Queries, pages, clicks, impressions, CTR, position |
get_performance_overview |
Site-level totals for a period |
compare_periods |
Side-by-side comparison of two date ranges |
get_advanced_search_analytics |
Analytics with dimension filters (country, device, etc.) |
get_search_by_page |
Queries driving traffic to a specific page |
URL inspection
| Tool | What it does |
|---|---|
inspect_url |
Indexing status, crawl date, mobile usability, rich results |
batch_inspect_urls |
Inspect up to 10 URLs at once |
check_indexing_issues |
Prioritised issue summary across multiple URLs |
Sitemaps
| Tool | What it does |
|---|---|
list_sitemaps |
All submitted sitemaps with status |
get_sitemap |
Details for a specific sitemap |
submit_sitemap |
Submit a new sitemap (requires GSC_ALLOW_DESTRUCTIVE=true) |
delete_sitemap |
Remove a sitemap (requires GSC_ALLOW_DESTRUCTIVE=true) |
Environment variables
| Variable | Default | Description |
|---|---|---|
GSC_ACCOUNTS_CONFIG |
~/.config/mcp-search-console/accounts.json |
Path to your accounts config |
GSC_ALLOW_DESTRUCTIVE |
unset | Set to true to enable sitemap submit/delete |
MCP_TRANSPORT |
stdio |
Set to sse for remote/Docker deployment |
MCP_HOST |
127.0.0.1 |
SSE bind host (use 0.0.0.0 for all interfaces) |
MCP_PORT |
3001 |
SSE bind port |
Remote deployment (Docker / VPS)
docker build -t mcp-search-console .
docker run \
-e MCP_TRANSPORT=sse \
-e MCP_HOST=0.0.0.0 \
-e MCP_PORT=3001 \
-e GSC_ACCOUNTS_CONFIG=/config/accounts.json \
-v /path/to/config:/config \
-p 3001:3001 \
mcp-search-console
Your MCP client connects to http://your-server:3001/sse.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.