LLM 與 Agent

Transform app ideas into shippable MVPs using rapid validation methodology. Use when the user wants to build a new app, evaluate an idea, generate a PRD, create Claude Code prompts, scope an MVP, or make product decisions. Helps avoid feature creep and over-engineering by focusing on core user loops and real user validation.

Develop and optimize AI prompt strategies for code generation. Use when creating new AI model strategies, optimizing prompts for Claude/GPT/Gemini/DeepSeek/Grok, or improving prompt quality. Includes model-specific optimization techniques.

ファイルアップロード、S3、ファイルシステム操作を扱う際に使用。

Generates consistent, professional cover images for business/fiscal articles using Ideogram with standardized prompts and naming conventions. Use when creating new articles, updating missing covers, or maintaining visual consistency across the content library.

Self-hosted GitHub Actions runner management via Podman Quadlet. Supportsmulti-instance pools with ephemeral storage, automatic token generation,and rolling updates. Use when users need to set up CI/CD runners fortheir GitHub repositories.

End-of-day cleanup and maintenance routine. Calculates Sleep Score (0-100) based on uncommitted work, branch cleanup, issue updates, CI status, and service shutdown. Automatically captures session state to STATUS.json. Use at end of work day before closing Claude Code.

Strategies for handling subagent failures with retry logic and escalation patterns.

MUST USE this skill when creating prompts for AI agents, structuring constraints, or designing prompt workflows. Secondary: implementing chain-of-thought, using example-based learning, or building meta-prompts.

Develop Notion templates and databases using Notion MCP tools in Claude Code. Orchestrates escape room design skills (narrative-architect, puzzle-designer, formula-master, localizer, playtester) and implements their output via Notion MCP API. Use when creating Notion templates, building databases, implementing game mechanics, or managing complex multi-step Notion development workflows. Handles rate limits, session persistence via Serena MCP, and iterative development cycles.

Strategic automation architecture advisor. Use when users want to plan automation solutions, evaluate their tech stack (Shopify, Zoho, HubSpot, etc.), decide between n8n vs Python/Claude Code, or need guidance on production-ready automation design. Invokes plan mode for complex architectural decisions.

Expertise in automated testing, code review practices, and quality standards enforcement. Activates when working with "lint", "test", "review", "coverage", "quality", "standards", or test automation.

Execute multiple Claude Code agents in parallel using the cpo CLI tool. Use when running parallel tasks, monitoring execution, or understanding the execution workflow.

Use this whenever a user wants to add a new feature or explitly states to research a feature/API or building a plan for a new feature. It itnerviews the user for feature details (if not provided), research the best API/service for their needs, confirm choice, then gather all implementation notes for their request and save them as a .claude/plans file.

Integrate Cloudflare services with Dokploy templates: R2 storage, DNS challenge for SSL, Zero Trust Access, Workers, WAF, and Tunnel. Default to CF services for external dependencies.

將提示詞和程式碼從 Claude Sonnet 4.0、Sonnet 4.5 或 Opus 4.1 遷移至 Opus 4.5。當使用者想要更新其程式碼庫、提示詞或 API 呼叫以使用 Opus 4.5 時使用。處理模型字串更新和針對已知 Opus 4.5 行為差異的提示詞調整。不會遷移 Haiku 4.5。

LlamaIndex Wolfram Alpha tool for computational knowledge queries, math solving, scientific calculations, and agent integration. Triggers: wolfram alpha, computational query, math solver, scientific calculation, WolframAlphaToolSpec.

Automatically analyze test coverage when user asks which code is tested, mentions coverage gaps, or shows code asking about testing. Identifies untested code paths and suggests test additions. Invoke when user asks "what's not tested?", "coverage", "untested code", or "which tests are missing?".

OpenAI Codex CLI integration

Guidelines for creating new Agent Skills and MCP tools for this WordPress MCP server. Use when adding new capabilities, creating skills, or registering MCP tools.

Comprehensive testing and verification checklist for agent-created pull requests. Use when preparing to submit a PR, verifying changes work correctly, or ensuring all evidence requirements are met. Covers Docker setup, database seeding, performance baselines, and proof-of-functionality requirements.