AI Services Market › Use cases › Sales, Marketing & Support
Sales, Marketing & SupportDone with ChatGPT 2026-10-27
Find technical SEO issues and generate code fixes
The author tested five SEO agents; one browsed a site, found broken canonicals, orphaned pages and missing schema, ranked them by impact and produced code fixes, while two others pulled Search Console data.
- What They Did
- The team gave five SEO agents the same brief on three live client sites: run a technical crawl, rank issues by impact, and produce shippable output. One agent browsed the site by itself, found broken canonicals, orphaned pages and missing schema, ranked them, and wrote code snippets with named file changes. It also compared pages against the top-ranking results to find topical gaps.
- How It Went
- It was the only genuine agent of the five and worked as a useful research assistant. But it recommended changes that would have broken the site's navigation, and it ranked issues by technical severity with no regard for business goals.
- Worth Knowing
- Review every fix before shipping. The agent missed context a human would spot in seconds, and the authors say it would eventually cause damage if left to execute unreviewed.
Try it yourself with ChatGPT
Review the technical SEO of [my site URL]. Browse the site and look for broken canonicals, orphaned pages, missing structured data, and redirect or indexing problems. Rank what you find by likely impact on search traffic, and for the top [number] issues give me a specific code fix I can paste in. Finish when I have the ranked list and the fixes, and do not edit the site yourself.