Slipstream: a search and AI-visibility scoring engine, checked against an expert

What it is

Slipstream is a white-label search, generative-engine and answer-engine workspace. It pairs Common Ground's audit method for search and AI readiness with an automated platform from Zen Aegis that re-checks sites on a schedule and scores them.

The decision

Zen Aegis would not put the engine's scores in front of a client until it had been tested against an expert doing the same audit by hand, and every difference between the two was explained.

What was built

• Built monitoring that re-crawls each site on a schedule, compares it with the last known good version and raises an alert when something changes.
• Built a scoring engine of 53 checks across four weighted categories, with quoted evidence stored for every result, critical failures that cap their category, and scores that are versioned and reproducible.
• Built a dashboard across every site in the fleet.
• On August 10, 2026 ran the engine against an expert hand audit of ten sites and sixteen check types.

Results

• 92.8 percent agreement with the hand audit and zero factual conflicts. All eleven differences were explained: three were grading scale, one site had changed, and seven were subpages the engine read that the manual audit had not.
• Run on Common Ground's own sites, it caught drift nobody had seen: a robots file pointing at a sitemap on a dead domain, a meta description reverted to a placeholder, and a placeholder analytics ID.
• The full method runs about 98 points with the expert layer added to the 53 automated checks. Listed on September 29, 2026 as ready to show clients.

Stories from this work

At a glance