Skill UI

Build production-grade procedural meshes in Three.js: profile extrusion, lofts, revolve, sweep, solidify, bevels and apertures, plus topology, coplanar, clash and semantic quality gates for hard-surface assemblies and humanoid robots.

This skill enables creating procedural vegetation such as trees, grass, and flowers within Three.js or WebGPU environments. It focuses on growth hierarchies, surface-following vines, and deterministic geometry generation. Techniques include GPU culling, wind simulation, and optimized rendering for virtual fields, ensuring high performance and visual fidelity in 3D scenes.

Routes Three.js graphics tasks to the smallest expert skill set, covering cameras, materials, geometry, atmosphere, shadows, temporal effects, and final image treatment.

Build high-quality 3D scenes, WebGL experiences, and interactive visual effects with Three.js. Covers scene setup, geometry, materials, lighting, animation, particles, and mouse-driven interaction for web-based 3D graphics.

Deploy and tune Web Application Firewalls (AWS WAF, Cloudflare WAF, ModSecurity) with managed rules for OWASP Top 10 protection, rate limiting, and geo-blocking.

This module adopts the persona, vocabulary, and critical pedagogical style of Yann LeCun. It provides deep insights into open-source technology, the concept of technological sovereignty, and the geopolitical incentives driving major AI companies (Meta, OpenAI, Google). It structures explanations using Socratic methods, grounding complex AI concepts in physical analogies.

This skill enables users to generate logo concepts from brand names or reference images using the official AI Logo Maker package on the Beatra service. It verifies vector geometry and trademark rights before finalizing raster generations. The tool requires explicit approval for paid hosted work and integrates with AI models like Claude or Codex. It ensures security through pinned verified packages without self-updating code.

Work a site's SEO and AI-visibility (GEO) backlog through the hosted LogNorm MCP server: audits, fixes, content, AI-answer tracking.

Geolocates and chronolocates photos with tool-verified reasoning: metadata and OCR intake, reverse image search, candidate ranking, satellite and street-view matching, sun and shadow math. Outputs coordinates with error radius, evidence images and tiered confidence.

A sparse-clue OSINT investigation methodology for extracting overlooked leads, connecting fragmented evidence, and testing explanations across sources. Use for multi-step CTF challenges, geolocation, event reconstruction, account verification, artifact interpretation, and infrastructure research.