В каталоге 182 820 навыков в 14 категориях.
After scaffolded practice, run an unassisted check — a problem with no AI help. Separates what the learner can do with s...
Detects when Claude is uncertain but presenting answers confidently. Teaches Claude to recognize and communicate uncerta...
Use whenever a figure or sentence shows an estimate, forecast, sample-based rate, model metric or average over seeds: de...
Route tasks to small model by default, escalate to large model only on low confidence detection, achieving 87% faster le...
Detect user's intent from their message and activate the right workflow silently.
Unified Consciousness Framework v4.0.0. Orchestrator architecture with K.I.R.A. Language System (6 modules), TRIAD unloc...
Expert skill for CUDA Unified Memory and memory prefetching optimization. Configure managed memory allocations, implemen...
Write the routing rules of the engine bridges into .unikit/bridge-routing.md: which bridge a kind of work goes through,...
A standardized CLI wrapper for Uni-Mol molecular ML workflows that handles representation extraction (embeddings), model...
Unit economics analysis, CAC LTV estimation, churn inference, gross margin per unit, customer economics, subscription ec...
Unity 6 AI and navigation guide. Use when working with NavMesh, pathfinding, NavMeshAgent, NavMeshSurface, NavMeshObstac...
Universal CapTP Derivation Skill
A prompt requesting an enhanced, more dynamic rendition of an existing visual while preserving its core composition, lig...
实现前走完四个象限，通过盲点扫描、头脑风暴、原型和访谈，交付一份 unknowns map。
Humanize LLM output so it reads like a careful human wrote it. Subtracts AI-isms (sycophancy, tricolons, em-dash overuse...
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
Unsloth-core provides the foundational framework for accelerating Large Language Model (LLM) fine-tuning. It leverages s...
Unsloth-cpt provides specific optimizations for Continued Pretraining (CPT) and domain adaptation. It addresses the crit...
Unsloth-datasets provides tools to prepare and optimize data for fine-tuning. Key features include standardizing externa...
Direct Preference Optimization (DPO) in Unsloth provides a way to align models with human preferences using paired data...
Unsloth provides a streamlined method to export fine-tuned models directly to GGUF format. It features "Dynamic 2.0" qua...
Unsloth-grpo enables training of reasoning models using Group Relative Policy Optimization (GRPO). This technique replac...
Unsloth models can be deployed using native optimized inference or through production serving engines like vLLM and SGLa...
Unsloth enables training on extreme context lengths (up to 89K+ on a single 80GB GPU) by utilizing manually derived Trit...
Unsloth supports a wide range of state-of-the-art model architectures, providing pre-quantized Hub variants and optimize...
Unsloth-orpo facilitates one-step preference alignment using Odds Ratio Preference Optimization (ORPO). Unlike DPO, whic...
Unsloth utilizes advanced quantization techniques to reduce the memory footprint of LLM fine-tuning. This includes "Dyna...
Supervised fine-tuning using SFTTrainer, instruction formatting, and multi-turn dataset preparation with triggers like s...
Unsloth supports fine-tuning for Speech-to-Text (STT) models like OpenAI Whisper. By applying its optimized LoRA pipelin...
Fine-tune LLMs with Unsloth using GRPO or SFT. Supports FP8, vision models, mobile deployment, Docker, packing, GGUF exp...
Unsloth-tts brings the library's performance optimizations to speech synthesis models. Specifically optimized for the Ll...
Unsloth-vision provides optimized support for fine-tuning multimodal models like Llama 3.2 Vision and Qwen2.5 VL. It all...
Prüfung der Untersuchungshaft – dringender Tatverdacht und Haftgründe § 112 StPO (Flucht, Fluchtgefahr, Verdunkelungsgef...
Untervermietung und Gebrauchsüberlassung an Dritte im Wohnraummietrecht – Erlaubnisvorbehalt § 540 BGB und Sonderkündigu...
Use after unwind:start to orchestrate layer-by-layer analysis using specialist subagents
Replace time with color chain derivations via seed chaining
Self-inverse derivation patterns where ι∘ι = id for frame-invariant self
Update development memory (events.jsonl) based on commit metadata and diff analysis. Automatically tracks features, fixe...
Sync AgentOps skills.
Annotate an active work session with a note, scope change, or new finding. Auto-detects the active session, synthesizes...
Update a specific crate skill to latest version
Sync entire framework (Pillars, Rules, Workflow, Skills) in one command - meta-skill orchestrating all update-* skills....
Update .claude/HISTORY.md with entries for recent commits that changed agent infrastructure files. Use after committing...
Smart merge for CLAUDE.md and AGENTS.md instruction files.
Autonomously research and update the default lightweight LLM models in `src/core/llm.ts` using web search to ensure the...
Self-improve the playwriter skill: reflect on past playwriter mistakes and unclear docs from this session, then update /...
Sync skills between projects - bidirectional copy with version detection and clean mode. TRIGGER when: user wants to syn...
Discover and sync all toolkit-using projects with the latest skills. Use when skills are modified, after the post-commit...
Update installed Turbo skills from the local repo with a dynamic changelog, conflict resolution for customized skills, a...
Apply plugin knowledge base updates to an existing generated system. Consults the Ars Contexta research graph for method...
upgrade frontend dependencies, backend dependencies, or mise tools through parallel worker tasks
升级pi-agent-core和pi-ai到最新版本
Increase supplied video resolution and clarity while preserving timing, content, grain intent, and natural detail. Use f...
Upstash context management and serverless Redis/Kafka
Turn a framework defect the agent hit while running a Magpie skill into a fix PR against `apache/magpie`, one PR per def...
Apply one restrained urban-daylight documentary exposure and color correction to a supplied photograph while preserving...
Urlaubsansprüche nach dem BUrlG – Entstehung und Wartezeit §§ 1, 4 BUrlG, gesetzlicher Mindesturlaub § 3 BUrlG in Werkta...
Token usage tracking, cost estimation, and usage analytics
Design a usage-based pricing scheme that scales revenue with value without scaring customers away — the metric that trac...
Stop-and-summarize protocol for when the account usage limit is nearly exhausted.
Usage Optimization Skill
Answer questions about the AI SDK and help build AI-powered features. Use when developers: (1) Ask about AI SDK function...
Générez des spécifications d’icône et des invites d’image pour les icônes de cas d’utilisation dans le référentiel de bl...
How to use the coding agents installed on this machine (claude, codex, droid, grok, hermes, opencode, copilot, pi) as pl...
You are a PROJECT MANAGER who delegates implementation work to sub-agents using runSubagent Use when configuring AI agen...
Use when executing implementation plans with independent tasks in the current session
Generate a clean white Tailwind CDN report page from user content, optionally password-gate viewing via client-side decr...
Structured checkpoint format for requesting human input. When an agent needs a decision, it must stop, present context,...
Detect user's technical level from first messages. Adjust all output language accordingly.
Use when coordinating multiple AI agents in real-time - provides inter-agent messaging via MCP tools
Use when coordinating multiple AI agents in real-time - provides inter-agent messaging via Rust PTY wrapper with file-ba...
Use when starting a session or when you need to discover which skill applies to the current task. This is the meta-skill...
Explain AgentOps workflows.
Route AI/ML tasks to correct Yzmir pack - frameworks, training, RL, LLMs, architectures, production
Claude SDK development with Messages API, Tool Use, Extended Thinking, streaming, and prompt caching
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY...
Invoke Claude Code CLI from Python orchestrators and shell scripts. Use when asked to "spawn claude as subprocess", "aut...
Route data analysis and machine learning work to the right skill in this suite. Use when starting any analysis, modeling...
Use when building networks that grow, prune, or adapt topology during training. Routes to continual learning, gradient i...
Run AgentOps on the Gas City (gc) substrate; gc is a guided dependency, not wrapped by ao.
Use when project knowledge could help - queries Kodex topics and flags outdated information
LLM specialist router to prompt engineering, fine-tuning, RAG, evaluation, and safety skills.
MANDATORY protocol enforcing knowledge check before EVERY response - prevents explaining systems without reading docs, c...
The architecture selection router for CNNs, Transformers, RNNs, GANs, GNNs by data modality and constraints
OpenAI SDK development with GPT-5 family, Chat Completions, Responses API, embeddings, and tool calling. Use for AI-powe...
Call the OpenAI API/SDK — GPT-5 family, Chat Completions, Responses API, embeddings, tool calling.
Use when an Agent needs to create, resume, advance, block, revise, complete, or abandon a durable current execution plan...
Routes to appropriate PyTorch specialist skill based on symptoms and problem type
Provides sub agents important information on how to use skills
Use when starting any conversation to initialize skill matching, or when unsure which skill applies to a request. Handle...
Auto-loaded at session start via SessionStart hook. Teaches skill invocation protocol, tool selection rules (look-at for...
Use at session start and before every task - establishes mandatory skill discovery and usage protocols to ensure relevan...
Invoke before responding to any non-trivial task—enforces the 1% rule that if any skill has even a 1% chance of improvin...
Skills wiki intro - mandatory workflows, search tool, brainstorming triggers
Core skill activation protocol - establishes mandatory workflows for finding and applying skills before any task
MANDATORY skill for all agents. Establishes how to discover, load, and use skills from SKILL-INDEX.md. Also detects when...
Use when loading skills for complex tasks — provides guidelines for using skills effectively, not every message.