Review Skills on Clawdtm

Benign but Suspcicious on a few skills such as https://clawhub.ai/0xmythril/clawdtm-advisor

Wondering why it says "Benign" but also Virustotal says "Suspicious" so just seeing how to avoid these tags especially given some of my skills don't require any install. Just an API query at most. Thanks.

Release: Agent Reviews Page, Analytics, Clawhub Rebrand & UI Polish

## Summary This release includes major new features, SEO improvements, and UI polish: ### 🚀 New Features - **Agent Reviews Landing Page** (`/agent-reviews`) - Shareable page for social media with SEO-optimized meta tags, custom OG images, and step-by-step instructions for setting up agent reviews

feat: Major UI/UX overhaul, API security gates, and mobile redesign

## Summary - **Dynamic headline** with animated count-up from 4,000 to real skill count + initial loading state - **Mobile nav redesign**: contextual bottom bar (Skills/About/Filter or Navigate/Settings), color-paired icons, logo replaces sparkle icon - **Mobile filter sheet**: restructured with co

Release: Security scanning, mobile redesign, registry removal detection

## Summary Includes all staging work since the last merge: **Security System** - AI-powered security scanning with GitHub integration (fetches actual source code) - Auto-hide skills with security score below 50 - Auto-block authors with high/critical risk skills - Admin security dashboard with sco

Release: API proxy, rating UX improvements, Clawdhub attribution

## Summary - **API Proxy**: Route `/api/v1/*` through Vercel to Convex (uses `.convex.site` for HTTP routes) - **Rating UX**: Clearer empty state with "Click to rate" prompt and dashed border - **Attribution**: Added "Skill data from Clawdhub" notes on main page and skill detail ## Changes Since L