Guidance for implementing minimal GPT-2 inference in constrained environments (code golf challenges). This skill should be used when implementing neural network inference from scratch, parsing binary checkpoint formats, implementing BPE tokenization, or working on code golf challenges involving ML models. Covers verification strategies and common pitfalls for checkpoint parsing and model inference.
Creates summary for human review - and comprehensive implementation plan for LLM, complete code examples, and verification steps
Creates Skills for Claude. Use when users request creating/updating skills, need skill structure guidance, or mention extending Claude's capabilities through custom skills.
Install skills from github.com/oaustegard/claude-skills into /mnt/skills/user. Use when user mentions "install skills", "load skills", "add skills", "update skills", "refresh skills", or references a skill not currently installed.
Internal skill. Use cc10x-router for all development tasks.
Use when starting any conversation - establishes mandatory workflows for finding and using skills, including using Skill tool before announcing usage, following brainstorming before coding, and creating TodoWrite todos for checklists
In Codex interactive mode, call delegate_autopilot automatically for multi-step or cross-cutting requests (including research-only), otherwise work normally.
A minimal example skill demonstrating Claude Code skill structure
Run the case feedback script to submit feedback for the current conversation. Only use when invoked by case-feedback-agent.
Query and analyze Claude Code history joined with Git data using SQL. Use when user asks about their Claude conversations, productivity patterns, commit history correlation, session analytics, or wants to explore their coding history with SQL queries.
Automatic review gate that runs after brainstorming completes - spawns PM, Architect, Designer, Security, and CTO agents in parallel, iterates until all approve
Generate and improve prompts using best practices for OpenAI GPT-5 and other LLMs. Apply advanced techniques like chain-of-thought, few-shot prompting, and progressive disclosure.
Guide for generating and editing images using generative AI with the nanobanana CLI
Master Anthropic's prompt engineering techniques to generate new prompts or improve existing ones using best practices for Claude AI models.
This skill should be used when users want to route LLM requests to different AI providers (OpenAI, Grok/xAI, Groq, DeepSeek, OpenRouter) using SwiftOpenAI-CLI. Use this skill when users ask to "use grok", "ask grok", "use groq", "ask deepseek", or any similar request to query a specific LLM provider in agent mode.
KERNELフレームワーク：1000時間のプロンプトエンジニアリングから発見された6つの原則。精度340%向上、時間67%削減を実現するエンタープライズ品質のプロンプト設計手法。AIへの指示を書く際に使用。
Configure Claude Code to use alternative AI model providers by setting up custom API endpoints and authentication. Use this skill when users want to switch from default Anthropic models to providers like Z.AI, OpenRouter, or other custom endpoints with compatible APIs.
GPT-5/4o, Claude 4.5, Gemini 2.5/3, Grok 4 vision patterns for image analysis, document understanding, and visual QA. Use when implementing image captioning, document/chart analysis, or multi-image comparison.
CLIP, SigLIP 2, Voyage multimodal-3 patterns for image+text retrieval, cross-modal search, and multimodal document chunking. Use when building RAG with images, implementing visual search, or hybrid retrieval.
Multi-agent frameworks beyond LangGraph. CrewAI crews, Microsoft Agent Framework, OpenAI Agents SDK. Use when building multi-agent systems, choosing frameworks.
Comprehensive prompt engineering with Chain-of-Thought, few-shot learning, prompt versioning, and optimization. Use when designing prompts, improving accuracy, managing prompt lifecycle.
Use when designing agent system prompts, optimizing RAG retrieval, or when context is too expensive or slow. Reduces tokens while maintaining quality through strategic positioning and attention-aware design.
Agentic workflow patterns for autonomous LLM reasoning. Use when building ReAct agents, implementing reasoning loops, or creating LLMs that plan and execute multi-step tasks.
Use when creating or improving golden datasets for AI evaluation. Defines quality criteria, curation workflows, and multi-agent analysis patterns for test data.
