mcpField
U
Untapped Search Console

Untapped Search Console

Local
github.comBusiness & Marketing0Updated 6d ago

About Untapped Search Console

Untapped Search Console reads a site's queries, pages, and performance changes from Google Search Console. It helps site owners identify search traffic trends and opportunities using read-only data.

Key features

  1. 01View clicks and impressions
  2. 02Break down performance by query, page, country, device, or date
  3. 03Find queries close to page one
  4. 04Highlight working and invisible searches
  5. 05Compare periods and spot gains or losses

Use cases

  1. 01Review search clicks and impressions for a period
  2. 02Find queries ranking positions 4–20 without clicks
  3. 03Identify pages or queries that gained or lost traffic
  4. 04Inspect performance by device or country

MCP support

Exposes read-only Google sign-in, property lookup, performance, totals, opportunity analysis, and comparison tools to MCP clients.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository
Packages
mcpbuntapped-search-console-mcp-0.2.0.mcpb0.2.0

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Untapped Search Console

What people search to find your site, which searches are close to page one, and what moved since last month, read from your own Search Console. Free and read-only. Built by Find Untapped Growth.

Tools

ToolAnswers
google_login / google_login_status / google_logoutOne read-only Google sign-in, shared with every Untapped Google plugin
gsc_sitesYour Search Console properties
gsc_totalsThe site's true clicks and impressions for a window
gsc_performanceRows by query / page / country / device / date (up to 3 at once), with filters
gsc_whats_workingearning (protect) · close, position 4–20 with no clicks (where work pays) · top3_no_clicks (fix the title/snippet) · invisible (past 20)
gsc_compareThis window vs the previous one: gainers, losers, new, lost, and both windows' true totals

Windows end 3 days back by default, because Search Console's last ~2 days aren't final and a window ending today shows a fake dip.

Install

  • Claude Code: /plugin install untapped-search-console --marketplace CTF-Max/untapped-plugins (or, from a terminal: claude plugin marketplace add CTF-Max/untapped-plugins, then claude plugin install untapped-search-console@untapped-plugins).
  • Codex: codex plugin marketplace add CTF-Max/untapped-plugins, then codex plugin add untapped-search-console@untapped-plugins.
  • Needs Python 3.10+ on PATH as python. Standard library only: nothing to pip install.

Setup

Same Google sign-in as Untapped Index Watch: a Google Cloud Desktop app OAuth client with the Search Console API enabled (steps in Index Watch's README). Claude Code: choose that file in the plugin's settings ("Google OAuth client file"). Codex: setx UNTAPPED_GOOGLE_CLIENT_SECRET "C:\path\to\client_secret.json" and restart. Then ask your agent to run google_login. Signed in already through Index Watch? Nothing more to do.

What it connects to

Only Google: accounts.google.com and oauth2.googleapis.com to sign in, searchconsole.googleapis.com and www.googleapis.com for your Search Console data. No telemetry. Details: PRIVACY.md.

Limits

  • Google hides rare queries for privacy, so query rows never add up to the site total; gsc_totals gives the real one.
  • Position is an average of averages: directional only.
  • gsc_compare compares each window's top 1,000 rows, so "lost" can mean "fell below the top 1,000".

Development

python -m unittest discover -s untapped-search-console/tests   # offline, 7 tests

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.