Observed arrival · 2026-10-08
MissSkill Bench: AI Skills, Tested Side by Side
A public benchmark compares a model’s answer to the same prompt with and without an AI Skill loaded.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- People evaluating AI Skills for coding and agent workflows
- Worth noticing
- The site says each comparison uses a fresh environment and leaves the paired answers unscored for readers to inspect.
Field notes
The example prompt asks for a Google Cloud Logging query for the checkout-api Cloud Run service and shows both generated queries. The catalog also groups entries by workflow and displays model labels and counts for completed comparisons. The homepage says Skill versions are checked and each run uses a clean environment, but the extracted page does not explain the benchmark’s sampling or review procedure.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue