mcpField
U
Untapped Site Audit

Untapped Site Audit

Local
github.comBusiness & Marketing0Updated 5d ago
RepositoryVisit website

About Untapped Site Audit

Untapped Site Audit is an agent-run website auditing plugin for site owners and SEO practitioners. It follows an eight-step bottleneck-oriented review and includes read-only checks across sitemap pages.

Key features

  1. 01Eight-step bottleneck-oriented audit
  2. 02Page-level search and HTML checks
  3. 03Sitemap-wide duplicate and status checks
  4. 04Marks unmeasurable findings as UNKNOWN

Use cases

  1. 01Reviewing a site's SEO bottlenecks
  2. 02Finding duplicate titles across sitemap pages
  3. 03Checking redirects and indexing directives

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-site-audit-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 Site Audit

A website audit your agent runs in the order that finds the real bottleneck (indexed → seen → buyer phrases → on-page targeting → visitors → engaged → buyers), plus free on-page checks across a whole sitemap. Built by Find Untapped Growth.

What's inside

  • Skill site-audit: the 8-step audit. It uses Untapped Index Watch (step 1), Untapped Search Console (step 2) and Untapped SERP (step 3) when they're installed, and labels anything it can't measure UNKNOWN instead of guessing.
  • page_facts: one page as a search engine reads it: status, redirect chain, title, meta description, meta robots + X-Robots-Tag + robots.txt (Googlebot), canonical, H1/H2s, schema types, hreflang, word count, links, images without alt, and the problems found.
  • audit_pages: page_facts for every sitemap URL (or a list), plus site-wide problems: duplicate titles, H1s and descriptions; noindex, redirected or non-200 pages listed in the sitemap.

Both tools are free and read-only; they fetch pages one at a time like a polite crawler.

Install

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

Setup

Nothing else to set up. To audit a local preview (localhost, 192.168.x.x), set UNTAPPED_AUDIT_ALLOW_PRIVATE=1; private addresses are refused otherwise.

What it connects to

Only the website you ask it to audit: its pages, its sitemap and its robots.txt, fetched one at a time. Nothing is sent anywhere else; no telemetry. Details: PRIVACY.md.

Limits

  • Reads the HTML as served. A page built by JavaScript after loading can look empty here.
  • Title 30–60 and description 70–160 characters are guidelines, not rules: Google rewrites snippets anyway.
  • Default 25 pages per audit (up to 200).

Development

python -m unittest discover -s untapped-site-audit/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.