Observed arrival · 2026-11-09
Akula checks AI-agent skills before installation
Akula scans AI-agent skills for security risks and measures whether they help or hurt, model by model.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Developers evaluating third-party AI-agent skills
- Worth noticing
- Reports bind verdicts to a bundle hash; a changed file invalidates old measurements, and unmeasured results remain unknown.
Field notes
The submission flow accepts a skill and its accompanying scripts, hooks, and configs, then hashes the bundle; the page says nothing is executed during the initial scan. For effect testing, it describes paired runs with and without the skill in an isolated sandbox, with results reported per model. Security scans run on every submission, while effect measurements cover selected skills rather than every submission.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
⊠LoginAccess appeared gated
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue