Run DYNO Mapper from Claude: 75 Tools, One Conversation

What MCP is, in one paragraph

The Model Context Protocol is an open standard published by Anthropic in late 2024. It defines how AI agents securely connect to external tools and data. Think of it as a universal adapter: one server, many clients. A platform builds one MCP server, and every MCP-compatible AI client can use it.

The DYNO Mapper MCP Server is published on npm as @dynomapper/mcp-server. It exposes 75 tools that map directly to the DYNO Mapper Public API: projects, sitemaps, content inventory, SEO audits, web and PDF accessibility, analytics, keyword visibility, and assignments. The official documentation covers the full tool reference and installation.

Why this matters if you already use DYNO Mapper

The point isn’t AI inside DYNO Mapper. It’s DYNO Mapper inside your AI workflow. Three things change as a result.

Audits become conversational. Instead of clicking through a dashboard, you ask follow-up questions. “Show me the missing meta descriptions. Now group them by directory. Now sort by which directories I crawled most recently.” The agent does the iteration, you do the thinking.

Cross-tool workflows become natural. With other MCPs connected, your DYNO Mapper data can flow directly into the rest of your stack: drafted into client emails, summarized into Slack updates, exported into Google Sheets, attached to project briefs. The data doesn’t sit in DYNO Mapper waiting to be exported; it moves where you need it.

Onboarding gets faster. New team members or clients don’t need a 30-minute walkthrough of the dashboard to ask useful questions. They open Claude and start asking: “What’s the accessibility score for our main marketing site, and what are the top three things blocking it from a higher score?” The MCP handles the rest.

Six ways teams are using it

1. Agency new-client onboarding

The pitch deck for a prospect lands a lot harder when it starts with a real audit, not a guess. With the MCP, an agency strategist can fire off a crawl during a discovery call and have substantive findings ready by the follow-up meeting two days later.

“Create a new DYNO Mapper project for acmecorp.com, wait until the crawl finishes, then pull the V.A.C.Q. score and a summary of the highest-severity SEO issues. Format the output as a one-page pitch addendum.”

Under the hood, the agent chains create_project_from_url, polls get_project_url_job_status until the crawl completes, then calls get_project_vacq_score and get_audit_results. The strategist gets a structured summary without writing a single dashboard query.

2. WCAG and ADA accessibility consulting

Accessibility audits don’t end at “here are the issues.” Clients need prioritization: which fixes matter most, which pages carry the most traffic risk, which problems cluster. The DYNO Mapper MCP lets accessibility consultants combine audit data with traffic data in a single prompt.

“For project 412, find the pages with WCAG Level A failures, then cross-reference with the highest-traffic pages from the last 30 days, and give me the prioritized fix list.”

The agent runs run_accessibility_checker, pulls get_accessibility_report_problems filtered by severity, fetches get_analytics_project_pages for traffic data, and joins the two. What used to be a three-tab manual reconciliation becomes one request.

3. PDF accessibility for regulated industries

This one is often overlooked. Government, healthcare, and education sites frequently host thousands of PDFs, each subject to Section 508 and WCAG compliance. Surfacing the non-compliant ones across a 10,000-page portal is a real problem, and the MCP solves it directly.

“List every PDF on the project with accessibility failures, grouped by the guideline they fail, with the file URL and severity for each.”

Tools used: get_accessibility_pdf_files, get_accessibility_pdf_guidelines, and get_accessibility_pdf_item_result. The output goes straight into a remediation ticket queue.

4. SEO triage at scale

The audit module exposes 12 dedicated tools for individual issue types: broken links, client errors, duplicate or missing or overly long titles and descriptions, non-crawlable pages, non-indexable pages, orphan pages. That granularity matters. It means the agent can pull exactly the slice that matters and ignore the rest.

“For project 412, list every page that has both a duplicate title AND a 4xx client error. Those are my critical fixes for this week.”

The agent fetches get_audit_duplicate_title and get_audit_client_error, computes the intersection, and returns just the pages that fail both checks. Triage that used to live in a spreadsheet now lives in the conversation.

5. Site migration planning

Migrations are where audits earn their keep. Before flipping DNS, you need a fresh inventory, a clean orphan-page list, and confidence that nothing important is about to be 404’d. The MCP makes pre-migration sweeps repeatable.

“Refresh project 256 to get the latest inventory, then compare it to last week’s snapshot. Flag any pages that have lost inbound links or moved out of the navigation.”

The agent runs refresh_project_url, polls get_project_url_job_status, then pulls get_inventory and get_inventory_link_occurrences to find pages whose link counts have dropped, and get_audit_orphan_pages to catch anything orphaned by recent content changes. Pair this with a sitemap export and you have your redirect plan.

6. Keyword strategy and reporting

Visibility tracking has its own ten-tool module covering keyword groups, daily rankings, and locale-specific results. The reporting use case is obvious (weekly client updates, monthly trend reviews), but the strategy use case is where the MCP gets interesting.

“Group these 40 keywords into three buckets (branded, competitor, generic), chart the weekly rank trends for each bucket over the last 90 days, and tell me which bucket has lost the most ground.”

Tools: create_visibility_keyword_group, create_visibility_keywords, update_visibility_keywords_group, get_visibility_project_ranking, get_visibility_keyword_ranking. The agent handles the bucketing logic and the rollup math; you handle the strategic call about where to invest.

Where it gets really interesting: combining MCPs

The use cases above all live inside DYNO Mapper. The real compounding happens when DYNO Mapper sits alongside other MCPs your team uses.

DYNO Mapper plus a WordPress MCP turns audit findings into one-step fixes: “Find every published page with a missing meta description, draft a description based on the page content, and update it in WordPress.”

DYNO Mapper plus a Google Drive or Dropbox MCP makes monthly client reporting almost free: “Generate this month’s audit summary for every client project, save each one as a PDF in the client’s shared folder, and post a Slack notification when done.”

DYNO Mapper plus a Gmail MCP closes the loop on outreach: “Find the broken outbound links on project 88, look up the new URLs for the destinations where I can, and draft outreach emails to the site owners for the rest.”

None of these workflows require custom code, integration platforms, or scheduled jobs. They’re conversations. The MCP standard means every tool your agent gains becomes available to every other tool. Your DYNO Mapper audits get more useful every time you connect a new service.

Getting started

Three steps.

1. Generate a Personal Access Token. Log in to DYNO Mapper and visit app.dynomapper.com/developer. In the Personal Access Tokens section, click Generate, name your token, and copy it.

2. Add the server to your AI client. For Claude Desktop, open the config file (~/Library/Application\ Support/Claude/claude_desktop_config.json on macOS, %APPDATA%\Claude\claude_desktop_config.json on Windows) and add:

{
  "mcpServers": {
    "dynomapper": {
      "command": "npx",
      "args": ["-y", "@dynomapper/mcp-server"],
      "env": {
        "MCP_API_KEY": "your_personal_access_token_here"
      }
    }
  }
}

The same configuration works in Cursor, Windsurf, and other MCP-compatible clients.

3. Restart, then ask. All 75 tools appear in your client’s tool list. Start with something simple (“What DYNO Mapper projects do I have?”) and build from there.

The full documentation covers every tool, the response format, and the architecture in detail.

The takeaway

The teams winning right now aren’t the ones with the biggest tool stack. They’re the ones who can chain their tools together fastest, with the least friction. MCP is what makes that chaining work: one protocol, many clients, every server reusable across every workflow.

DYNO Mapper now sits on that chain. Seventy-five tools, one conversation. The next audit you run could be a sentence away.

Read the documentation →