Loading...
Loading...
Found 5,561 Skills
Calibrate an LLM judge against human labels using data splits, TPR/TNR, and bias correction. Use after writing a judge prompt (write-judge-prompt) when you need to verify alignment before trusting its outputs. Do NOT use for code-based evaluators (those are deterministic; test with standard unit tests).
Create diverse synthetic test inputs for LLM pipeline evaluation using dimension-based tuple generation. Use when bootstrapping an eval dataset, when real user data is sparse, or when stress-testing specific failure hypotheses. Do NOT use when you already have 100+ representative real traces (use stratified sampling instead), or when the task is collecting production logs.
Expert blueprint for RPG stat systems (attributes, leveling, modifiers, damage formulas) using Resource-based stats, stackable modifiers, and derived stat calculations. Use when implementing character progression OR equipment/buff systems. Keywords stats, attributes, leveling, modifiers, CharacterStats, derived stats, damage calculation, XP.
Expert blueprint for rhythm games including audio synchronization (BPM conductor, latency compensation with AudioServer.get_time_since_last_mix), note highways (scroll speed, timing windows), judgment systems (Perfect/Great/Good/Bad/Miss), scoring with combo multipliers, input processing (lane-based, hold note detection), and chart/beatmap loading. Based on DDR/osu!/Beat Saber research. Trigger keywords: rhythm_game, audio_sync, timing_judgment, note_highway, combo_system, BPM_conductor, latency_compensation.
Angular 12-16 performance optimization guidelines with NgModules, RxJS patterns, and classic template syntax (*ngIf, *ngFor). Use when maintaining or working with pre-v17 Angular codebases. Triggers on tasks involving Angular components, services, modules, or performance improvements in legacy projects.
Global Agent rules, including language, response style, debugging priority, engineering quality baseline, mandatory code metric limits, security baseline, test verification standards and Skills routing table. Applicable to all programming tasks.
Evidence-based memory optimization from real usage patterns. Analyzes recall performance, identifies bottlenecks, suggests consolidation/pruning/enrichment, and tracks improvement over time via checkpoint Q&A.
Cognitive science-based deep source code understanding assistant (Chinese improved version). Supports three analysis modes: Quick (overview), Standard (comprehension), Deep (mastery, automatically uses parallel processing for large projects). Integrates elaborative interrogation, self-explanation testing, and retrieval practice to help truly understand and master code.
Build an Opportunity Solution Tree (OST) to structure product discovery — map a desired outcome to opportunities, solutions, and experiments. Based on Teresa Torres' Continuous Discovery Habits. Use when structuring discovery work, mapping opportunities to solutions, or deciding what to build next.
Fast, lightweight 2D rendering engine for creating interactive graphics, particle effects, and canvas-based applications using WebGL/WebGPU. Use this skill when building 2D games, particle systems, interactive canvases, sprite animations, or UI overlays on 3D scenes. Triggers on tasks involving PixiJS, 2D rendering, sprite sheets, particle effects, filters, or high-performance canvas graphics. Alternative to Canvas2D with WebGL acceleration for rendering thousands of sprites at 60 FPS.
Smart answering framework that automatically adapts to question types and delivers evidence-based, natural responses.
Audit and prune bloated CLAUDE.md or AGENTS.md context files using evidence-based criteria from research on what actually helps coding agents. Use when a user asks to trim, audit, review, or improve their CLAUDE.md, AGENTS.md, or any repository context file for AI coding agents.