designesy.org/report3 engines67 checks8 synthesis checkscontract 0.1.1 provisional
Report
One URL through three engines, weighted into one grade you can cite: the 45-check contract score 50%, drift 30%, AI readiness 20%.
The URL, then three engines in parallel.
The combined score, its grade, and a coherence check.
Build a report and each share of the bar fills to its engine's score. The combined score is their weighted sum.
- Contract score 50%
- The broadest measure: tokens, type, motion, accessibility, identity, runtime.
- Drift radar 30%
- Whether the system holds together as AI-written code piles up.
- AI readiness 20%
- What an agent can read about the system before it builds.
If an engine returns nothing, the others are re-weighted and the synthesis checks say so.
- pass
- needs work
- fail
- does not apply
Method
Fire three engines
The contract score, the drift radar and AI readiness, in parallel, each fetching the page itself.
Weigh them
Contract score counts 50%, drift 30%, readiness 20%. If one returns nothing, the others are re-weighted.
Check the report
Eight synthesis checks confirm every engine ran and that no engine sits more than 30 points from the combined score.
Grade
The weighted sum, rounded, takes the same letter scale as every engine. A is 90 and up.
combined score = round(contract score × 0.5 + drift × 0.3 + readiness × 0.2)A ≥ 90B ≥ 80C ≥ 70D ≥ 60F belowThe synthesis checks grade the report itself, apart from the combined score.