---
title: Dualmark Md Companion
category: product
entity_type: skill
price: $15
canonical: https://forgehouse.ai/skills/dualmark-md-companion/
lang: en
hreflang_alt: https://forgehouse.ai/tr/skiller/dualmark-md-companion/
last_updated: 2026-06-20
---

# Dualmark Md Companion

> Implement dualmark-style .md companion endpoints alongside HTML pages for AEO/GEO every URL…

Builds a markdown twin for every HTML page so AI search engines like ChatGPT, Claude, and Perplexity get clean, parseable content instead of HTML clutter. It integrates the dualmark library into Astro 5 and Next.js 15 via HTTP content negotiation, with cache invalidation, content-drift detection, and AI-crawler discovery. The foundation for getting your pages cited by AI answer engines.

## Use cases
- Serving a .md companion endpoint per page for AEO/GEO
- Fixing 'ChatGPT cites my competitor, not me' by removing HTML clutter for bots
- Letting AI bots cleanly consume a Sanity CMS or MDX blog
- Setting up llms.txt and rel=alternate discovery in Next.js 15 or Astro 5
- Allowlisting AI crawlers in robots.txt and ai.txt
- Generating markdown twins for programmatic/pSEO pages at scale

## Benefits
- Increase your odds of being cited by AI answer engines, not just your competitors
- Feed AI crawlers clean markdown instead of JS-heavy HTML soup
- Avoid duplicate-content penalties with proper noindex and discovery headers
- Keep HTML and markdown perfectly in sync with single-source rendering

## What’s included
- Astro 5 and Next.js 15 integration configs with content negotiation
- Safe HTML-to-Markdown converter pipeline with three-layer XSS defense
- Cache invalidation that purges HTML, markdown, and llms.txt together
- Four-layer AI-crawler discovery: robots.txt, ai.txt, llms.txt, and rel=alternate
- Content-drift detection comparing HTML and markdown word counts daily
- Conformance verification scoring and an eight-step implementation workflow

## Who it’s for
Marketing engineers and content teams who want their site cited by AI answer engines and need clean markdown infrastructure for AI crawlers.

## How it runs
AI crawlers read markdown better than HTML, so every page gets a twin. The skill audits your conformance baseline, wires content negotiation into your stack, and watches daily for drift between the two versions.
1. Audits the current site with the dualmark conformance check to get the 0-125 baseline (most sites start in the 20-40 range).
2. Installs the adapter for your stack (Astro 5, Next.js 15 or Cloudflare Workers) and wires the content negotiation router: AI bot User-Agents and Accept: text/markdown requests get the markdown twin, browsers get HTML plus a rel=alternate link header.
3. Maps each content collection (blog, pricing, compare, glossary) to a converter that renders the markdown from the same single CMS source as the HTML, because two separate fetches is exactly how content drift starts.
4. Generates the discovery layer: an auto-built llms.txt index, an ai.txt policy file, the robots.txt AI bot allowlist, and a link rel=alternate tag in every HTML head.
5. Ties cache invalidation together so one CMS publish purges the HTML page, its .md twin and llms.txt atomically, origin cache first, edge cache second.
6. Verifies conformance at 80+ out of 125 and installs a daily drift cron that compares HTML and markdown word counts; a 20%+ delta raises an alarm before AI bots read stale content.

## FAQ
### I am on Next, not Astro, and my content is in Sanity. Does that work?
It integrates with both Astro 5 and Next, and pulls from sources like Sanity CMS or MDX. If you are on one of those, the markdown twin fits your existing pipeline.

### Does a markdown copy of my page really change whether an AI engine cites me?
AI bots parse clean markdown far more reliably than cluttered HTML, so your content reaches them intact instead of half-read. It removes a real barrier, the bot sees what you actually wrote.

### Will this guarantee ChatGPT cites me instead of a competitor?
No tool can guarantee a citation, that depends on your content quality and authority too. What this fixes is the infrastructure problem: making sure bots can cleanly read you in the first place.

## Price
$15, one-time, no subscription. VAT included.

Related guide: [How to automate SEO and AEO with Claude](https://forgehouse.ai/guides/automate-seo-claude/)