Menu
Skills in this repository
SkillsMP has collected 5,417 skills from majiayu000/claude-skill-registry. Open a skill to review its source and details.
majiayu000/claude-skill-registryShowing 40 of 5,417 collected skills.
Generate self-contained collaboration documents for sharing issues with external AI systems (Gemini, ChatGPT, etc.) with structural anti-skip enforcement (Execute-Verify-Record pattern at every step). Interactively gathers context, reads actual code files,…
Provides comprehensive guidance for Spring AI Alibaba including Alibaba Cloud AI services integration, model APIs, and AI application development. Use when the user asks about Spring AI Alibaba, needs to use Alibaba Cloud AI services, or integrate AI…
Source text: Chinese
Model Context Protocol (MCP) server implementation patterns with Spring AI. Use when building MCP servers to extend AI capabilities with custom tools, resources, and prompt templates using Spring's official AI framework.
Provides comprehensive guidance for Spring AI including AI model integration, prompt templates, vector stores, and AI applications. Use when the user asks about Spring AI, needs to integrate AI models, implement RAG applications, or work with AI services in…
Source text: Chinese
This skill should be used when the user asks about "Effect Stream", "Stream.from", "Stream.map", "Stream.filter", "Stream.run", "streaming data", "async iteration", "Sink", "Channel", "Stream.concat", "Stream.merge", "backpressure", "Stream.fromIterable",…
Techniques for ensuring LLM responses adhere to strict JSON schemas, utilizing Pydantic models, JSON mode, and schema-based refusals. Triggers: structured-output, pydantic, json-schema, json-mode, llm-response-parsing.
次のタスクの内容（コーディング、リサーチ、リファクタリングなど）に応じて、最適なLLMモデルを提案する
Source text: Japanese
Provides flag definitions and usage rules when double-dash flags (--flag-name) detected in user prompt - describes execution modes, tool selection, and behavioral modifications
Expert SuperClaude prompt engineering assistant that analyzes user needs and crafts optimal prompts using the full SuperClaude framework - commands, flags, personas, MCP servers, wave orchestration, parallel execution patterns, continuous execution…
タスクの複雑さに応じて Claude のモデルを自動的に切り替える。複雑なタスク（アーキテクチャ設計、大規模リファクタリング、難解なバグ修正）では opus を、単純なタスク（小さな修正、フォーマット、簡単な質問）では sonnet を自動選択。コストと品質のバランスを最適化。
Source text: Japanese
Build neuro-symbolic LLM applications with Synalinks framework. Use when working with DataModel, Program, Generator, Module, training LLM pipelines, in-context learning, structured output, JSON operators, Branch/Decision control flow, FunctionCallingAgent,…
System prompt assembly for blah.chat AI backend. Multi-layer prompt construction with priority ordering, parallel context loading, memory truncation, budget awareness. Use when working with "system prompt", "base prompt", "identity memories", "custom…
Parse natural language task input and extract structured data (title, due date, priority, category) using Google Gemini AI
使用Tavily API进行网络搜索，获取实时信息、回答问题或研究主题
Source text: Chinese
Web search, online search, real-time search, internet search, Google alternative, Bing alternative, DuckDuckGo alternative, search the web, lookup online, find information, research,查询,搜索,搜索结果,网页搜索,联网搜索,实时搜索,网络查询,资料查找,信息检索,最新资讯,新闻搜索, Tavily Search API for…
Use Tavily Search API for optimized, real-time web search results for RAG. A pre-configured, cost-effective search tool.
Source text: Mixed languages
"Tavily AI search API for LLM applications: web search, content extraction, site crawling, mapping, and research. Keywords: Tavily, AI search, RAG, web search API, LLM search, extract, crawl, map, research, tavily-python."
Executes Tavily AI search, extraction, crawling, mapping, and deep research via Python CLI. Use when web search, URL content extraction, site crawling, or multi-step research needed.
Web search, content extraction, crawling, and research capabilities using Tavily API. Use when you need to search the web for current information, extracting content from URLs, or crawling websites.
Step-by-step guidance for tensorrt LLM.
Deploy LLMs with Hugging Face Text Generation Inference. Configure quantization, continuous batching, and tensor parallelism. Use for production LLM serving, high-throughput inference, and model deployment.
CRITICAL - Read FIRST before any work. Strategies to minimize token usage and reduce costs by 60-80%.
Token usage monitoring and cost tracking with configurable thresholds. Reference this skill to understand pricing and limits.
Imported skill tools from langchain
Token-Oriented Object Notation (TOON) format expert for 30-60% token savings on structured data. Auto-applies to arrays with 5+ items, tables, logs, API responses, database results. Supports tabular, inline, and expanded formats with comma/tab/pipe…
Formats structured data using TOON v2.0 to minimize tokens while preserving readability. Use when outputs include tables, logs, events, or repeated records and token budgets matter. Triggers include "format table", "structured data", "TOON", "minimize…
TOON format knowledge and usage patterns for agent communication and memory persistence in plan-marshall marketplace
Voice-first runtime steering + nightly deep analysis to learn per-user conversation priors.
Specifies guidelines for the AI assistant to provide accurate, thoughtful answers, admit when it doesn't know something, and be concise while ensuring clarity. This rule promotes trustworthy and helpf
Best practices for using the Council MCP server in Tzurot v3 development - When to consult external AI, how to structure prompts, model selection, and multi-turn conversations. Use when planning major changes or needing a second opinion.
Route tasks to small model by default, escalate to large model only on low confidence detection, achieving 87% faster learning and 10-30x cost reduction while maintaining accuracy. Use for cost optimization, confidence-based delegation, routine vs complex…
Utilizing Dynamic 4-bit quantization, FP8 training, and 8-bit optimizers to minimize VRAM usage without sacrificing accuracy. Triggers: quantization, dynamic 4-bit, fp8, bitsandbytes, adamw_8bit, qat.
Supervised fine-tuning using SFTTrainer, instruction formatting, and multi-turn dataset preparation with triggers like sft, instruction tuning, chat templates, sharegpt, alpaca, conversation_extension, and SFTTrainer.
Fine-tuning Speech-to-Text models like Whisper using Unsloth's optimized LoRA pipeline. Triggers: stt, whisper, transcription, audio fine-tuning, speech-to-text, audio normalization.
Fine-tune LLMs with Unsloth using GRPO or SFT. Supports FP8, vision models, mobile deployment, Docker, packing, GGUF export. Use when: train with GRPO, fine-tune, reward functions, SFT training, FP8 training, vision fine-tuning, phone deployment, docker…
Fine-tuning Text-to-Speech (TTS) models with Unsloth for voice cloning and synthetic speech (triggers: TTS, text-to-speech, voice cloning, Orpheus-TTS, audio fine-tuning, speech synthesis).
Fine-tuning multimodal vision-language models (Llama 3.2 Vision, Qwen2.5 VL) using optimized vision layers (triggers: vision models, multimodal, Llama 3.2 Vision, Qwen2.5 VL, UnslothVisionDataCollator, finetune_vision_layers).
Usage Optimization Skill
Detect user's technical level from first messages. Adjust all output language accordingly.
LLM specialist router to prompt engineering, fine-tuning, RAG, evaluation, and safety skills.