Store important project context in Engram Cogitator's persistent memory. Use after making architectural decisions, discovering codebase gotchas or workarounds, identifying recurring patterns, or when user states preferences. Triggers: "remember this", "store this", "save for later", "don't forget", architectural decision, learned something, discovered a gotcha.
AIエージェント・システムロール・ペルソナのための役割プロンプト設計スキル。 責務分離、専門家思考様式の適用、効果的なプロンプト構造化の指針を提供する。 Anchors: • The Pragmatic Programmer (Andrew Hunt, David Thomas) / 適用: DRY原則・責務分離 / 目的: ロール設計の品質向上 • Domain-Driven Design (Eric Evans) / 適用: ユビキタス言語・境界づけられたコンテキスト / 目的: 責務境界の明確化 • Thinking, Fast and Slow (Daniel Kahneman) / 適用: 専門家思考様式の理解 / 目的: 適切な思考モード設計 Trigger: Use when designing AI agent roles, system prompts, persona definitions, or separating responsibilities between agents. role prompting, persona design, agent role, system prompt, responsibility separation, ロール設計, ペルソナ
Generate comprehensive model cards and upload fine-tuned models to Hugging Face Hub with professional documentation
Orchestrate parallel CLI agents (Claude Code, Codex, Gemini) for competitive evaluation. Use when user says "run multi-agent", "compare agents", "launch competitive evaluation", "use parallel agents", or complex tasks (>7/10) where multiple approaches exist and best solution matters.
Claude Codeのログ分析からSkillを自動生成するスキル。操作パターンを検出し、繰り返し作業をSkill化して自動化を支援する。
Add a new AI provider or model for recipe generation. Use when adding support for a new LLM provider (Anthropic, Google, etc.) or adding models to an existing provider.
Patterns for launching and managing parallel subagents efficiently.
Complete llama.cpp C/C++ API reference covering model loading, inference, text generation, embeddings, chat, tokenization, sampling, batching, KV cache, LoRA adapters, and state management. Triggers on: llama.cpp questions, LLM inference code, GGUF models, local AI/ML inference, C/C++ LLM integration, "how do I use llama.cpp", API function lookups, implementation questions, troubleshooting llama.cpp issues, and any llama-cpp or ggerganov/llama.cpp mentions.
Designs and optimizes prompts for large language models to achieve better, more consistent outputs. Trigger keywords: prompt, LLM, GPT, Claude, prompt engineering, AI prompts, few-shot, chain of thought.
プロンプトのテスト、評価、反復改善を専門とするスキル。A/Bテスト、評価メトリクス、自動化されたプロンプト品質保証により、本番環境で信頼性の高いプロンプトを実現します。 Anchors: • Test-Driven Development: By Example (Kent Beck) / 適用: Red-Green-Refactorサイクル / 目的: 反復的な品質改善 • LLM-as-a-Judge pattern / 適用: 自動評価とスコアリング / 目的: スケーラブルな品質評価 • A/B Testing for AI Systems / 適用: プロンプト比較実験設計 / 目的: データドリブンな改善 Trigger: Use when testing prompts, evaluating prompt quality, running A/B tests on prompts, implementing automated prompt evaluation, or establishing continuous prompt improvement cycles. Keywords: prompt testing, A/B testing, evaluation metrics, LLM-as-a-judge, prompt quality, automated evaluation, regression testing
Systematic prompt evaluation framework with MATH, GSM8K, and Game of 24 benchmarks. Use when evaluating prompt effectiveness on standard benchmarks, comparing meta-prompting strategies quantitatively, measuring prompt quality improvements, or validating categorical prompt optimizations against ground truth datasets.
Interactively configures Claudikins Automatic Context Manager settings by asking the user questions about trigger threshold, snooze duration, summary length, and other preferences. Use when the user wants to customize Claudikins Automatic Context Manager behavior or runs '/acm:config'.
Reference for configuring tool permissions when launching Claude Code agents. Use when setting up --allowedTools flags, restricting file access, or configuring agent permissions.
专业的LangGraph AI应用开发技能，提供从概念到生产的完整开发指导。基于Context7最新调研，涵盖StateGraph设计、多代理协作、RAG系统实现、生产部署等核心场景。使用此技能构建有状态、多参与者、长期运行的AI代理应用。
推論パターンを選択し、根拠サマリーを短く提示するためのプロンプト設計スキル。 自己一貫性の比較、推論パターンの適用、説明の明確化を通じて、再現性の高い結論を導く。 Anchors: • The Pragmatic Programmer / 適用: 手順設計 / 目的: 実践的な整理 • Reasoning and Logic / 適用: 推論パターン / 目的: 一貫性のある説明 • Self-Consistency (Wang et al.) / 適用: 複数案比較 / 目的: 精度向上 Trigger: Use when selecting reasoning patterns, designing prompts for structured explanations, or comparing multiple solution paths for higher confidence.
Export the current Pi session transcript to an HTML file for inclusion in a pull request. Use when the user is ready to submit a PR and wants to include an AI session transcript.
Modular orchestration of agent patterns from Anthropic's engineering guide. Intelligently selects and implements prompt chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer, and autonomous agents. Includes pattern combinations and language-specific implementations.
Generates implementation specifications from conversation context optionally enriched with GitHub issue data
Execute prompts from ./prompts/ directory with various AI models. Use when user asks to run a prompt, execute a task, delegate work to an AI model, run prompts in worktrees/tmux, or run prompts with verification loops.
