Conductor Website Monitoring to MCP: SEO Health, Issues, Pages, Alerts and Crawler Visits in any AI Agent

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Conductor Website Monitoring, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

Conductor Website Monitoring

Real-time SEO auditing data from Conductor Website Monitoring (formerly ContentKing): monitored websites and segments, health scores and issue statistics, tracked pages with their SEO data, SEO issues with affected pages, broken and redirected links, alerts and log file crawler visits, plus re-auditing a page on demand. Read-only token; re-auditing needs the CMS API token.

Setup

Setting up takes about 2 minutes. You need a Conductor Website Monitoring (ContentKing) account and access to its integration tokens, which an account admin can see.

1. In Conductor Website Monitoring open Account, the Account tab, Integration Tokens.
2. Copy the Reporting API token into CONDUCTOR_REPORTING_API_TOKEN. Paste the token only, without the word token: the connector sends it as Authorization: token <token>. This token is read-only.
3. Optional: copy the CMS API token from the same page into CONDUCTOR_CMS_API_TOKEN. Only conductor_monitoring_trigger_page_crawl needs it; without it that tool answers 401.
4. Install the connector and run conductor_monitoring_list_websites: it lists the monitored websites and confirms the token.

The connector uses the Reporting API v2 (https://api.cm.conductor.com). Most tools need a website_id: the id from conductor_monitoring_list_websites.

What it reads: the monitored websites with domain and page capacity, and their segments (groups of pages such as blog or product pages). Statistics of a website or segment: health score, number of issues, URL counts by type and status code, indexability, title, meta description and H1 distributions and Lighthouse scores, also for a past date with captured_at. Every tracked URL with its SEO data, and for one page everything known: SEO fields, status, indexability, Lighthouse, open issues, Schema.org data, content, crawl frequency from log files and Enrichment Fields. SEO issue types with how many pages are open, closed, ignored or not applicable, and the pages affected by one issue. The open issues of one page and its broken, redirected and canonicalized links. Active alerts (sudden changes such as pages becoming non-indexable, changed titles, robots.txt changes or 5xx errors) and the pages that entered or left them. Daily visits of search engine and AI crawlers (Googlebot, Bingbot, OpenAI, Anthropic, Perplexity, Meta) to up to 50 URLs, which needs the Log File Analysis feature; the default range is the last 30 days.

What it writes: only conductor_monitoring_trigger_page_crawl, which queues one URL for immediate auditing through the CMS API, so a change that was just published is checked now instead of at the next scheduled visit. The URL must belong to a website in the account. The AI then reads conductor_monitoring_get_page_issues a little later. No tool is switched off at install.

Data formats: URLs are full and absolute (https://www.example.com/page). Dates are ISO 8601; date ranges take YYYY-MM-DD. Health is a score from 0 to 1000. Issue names have the form <group>/<issue>, e.g. title/missing, meta_description/duplicate, h1/too_many, links/broken or lighthouse/lcp. Segments go into the scope filter as segment:<id> or segment_label:<label>.

Paging: per_page is 1 to 1000 results. conductor_monitoring_list_pages and the link tools return next_page_cursor, which the AI passes as page_cursor until it is missing; on sites with more than 10,000 pages that is faster than page. conductor_monitoring_list_affected_pages and conductor_monitoring_list_alert_pages page with page from 1.

Limits: the API allows 6 requests per second per IP address. Above that it answers 429 and blocks for a minute, so the AI does not loop over pages faster than needed. conductor_monitoring_get_metrics (a daily time series of a website) and conductor_monitoring_inspect_url (finds a URL without a website id) use the Reporting API v1, which Conductor no longer develops and plans to retire.

Common errors: 401 means a wrong or missing token, or a missing CMS API token for the re-audit tool. 422 means a malformed Authorization header, usually because the word token or quotes were pasted with it. 400 or 404 means an unknown website id or a URL that is not tracked.

The connector does not change website settings, segments or alerts, does not add websites, and does not mark issues as ignored or fixed: it reports, and the fix happens on your site.

Prerequisites: CONDUCTOR_REPORTING_API_TOKEN (optional: CONDUCTOR_CMS_API_TOKEN)

Install the connector locally

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Open http://localhost:3000/connectors/store, pick Conductor Website Monitoring and paste in the env vars listed under Prerequisites.

Available tools

Example prompts

• "What is the health score of www.example.com and how did it change compared with 30 days ago?"
• "Which pages have a missing or duplicate meta description? Group them by segment."
• "List the active alerts of our main website and the pages that entered each one."
• "Show the broken links on https://www.example.com/pricing with their anchor text."
• "How often did Googlebot and the AI crawlers visit our top 10 product pages last month?"

Next steps

• Connect to Claude
• Connect to ChatGPT