oh-my-kiro

Generate professional PDF documents with correct CJK (Chinese/Japanese/Korean) rendering. Trigger when user says 'generate PDF', 'create PDF', 'export PDF', '生成 PDF', '导出 PDF', '做个 PDF', 'make a PDF', or when any task requires producing a PDF file. Also trigger when user mentions PDF formatting issues, garbled text, or font problems.

Google Stitch design-to-code workflow with design system management. Trigger when user mentions Stitch, design-to-code, UI design, visual comparison, design.md, design system, ShadCN conversion, redesign from screenshot, or wants to pull designs from Stitch into local code. Also trigger when user says '做个页面', 'build a landing page', '用 Stitch 设计', 'convert design', 'export from Stitch', '设计系统', or uploads wireframes/screenshots for UI generation.

Full plan lifecycle: deep understanding → write plan with TDD checklist → parallel review → Ralph Loop execution. Trigger when user says 'plan', 'design', 'implement', 'build', 'architect', '@plan', '@execute', or describes a multi-step task that needs structured breakdown. Also trigger for feature requests, system redesigns, or migration projects.

Code and plan review with multi-angle dispatch. Trigger when user says 'review', 'code review', 'check my code', 'PR review', '@review', or when completing a plan phase that requires review. Also trigger before merge, after major feature completion, or when user asks for feedback on implementation quality.

Systematic debugging: reproduce → hypothesize → verify → fix. Trigger when encountering bugs, test failures, errors, crashes, unexpected behavior, or when user says 'debug', 'not working', 'broken', 'fails', 'error', 'exception', 'why does this happen', 'investigate'. Also trigger on stack traces, error logs, or non-zero exit codes.

Enforces coding best practices: deep-read before modify, LSP-first navigation, TDD red-green-refactor, minimal changes, self-review, verification. Trigger when writing code, modifying source files, fixing bugs, refactoring, optimizing code, entering a worktree or submodule, or when user says 'write code', 'implement', 'fix this', 'fix PR', 'add feature', 'add field', 'refactor', 'optimize', 'modify', 'change this', 'update code', 'apply feedback', 'apply review', '改代码', '修复', '加个', '改一下', '优化'. Also trigger when creating/editing .java .py .ts .js .go .rs .sh .tsx .jsx .vue .css .scss files, or after debugging identifies root cause and implementation begins.

Create high-quality, production-ready skills from scratch. Trigger when user says 'create skill', 'new skill', 'write a skill', 'make a skill', '创建 skill', '写个 skill', 'skillify', or when packaging a repeated workflow into a reusable skill. Also trigger when reviewing or improving existing skills.

Distill a top-level principle into knowledge/rules.md. Trigger when user says 'add rule', 'new principle', 'enforce this', '@agent', or wants to codify an architectural decision, workflow principle, or behavioral guideline into persistent rules.

Fetch current library/framework documentation via Context7. Trigger when user asks about libraries, frameworks, API references, code examples, setup/configuration, or mentions specific frameworks like React, Vue, Next.js, Prisma, Supabase, Express, Tailwind, Django, FastAPI, Spring Boot, etc. Also trigger when user says 'how to use', 'API docs', 'latest syntax', or needs code involving third-party packages.

Branch completion workflow: verify tests → present 4 options (merge/PR/keep/discard) → execute → cleanup worktree. Trigger when implementation is complete, all tests pass, user says 'done', 'finish', 'merge', 'create PR', 'push', 'wrap up', '@cpu', or when ready to integrate work back to main branch.

Capture knowledge into episodes.md. Trigger when user says 'remember', 'capture this', 'lesson learned', 'note this', '@know', or when preserving insights, corrections, or discoveries from the current conversation.

Multi-level research: built-in knowledge → web search → Tavily deep research API. Trigger when user says 'research', 'investigate', 'find out', 'compare', 'what is', 'how does X work', 'competitive analysis', 'market research', '@research', or needs information beyond the current codebase and knowledge base.

Agent self-learning: promote recurring episodes to rules, capture complex corrections. Trigger when hook outputs 🔥 or ⬆️ (promotion needed), 🚨 (complex correction), or when user says 'reflect', 'learn from this', 'remember this', 'always do X', 'never do Y', '@reflect'. Also trigger when same mistake pattern appears ≥3 times in episodes.

Evidence before claims — run verification commands before any completion/success statement. Trigger when about to claim 'done', 'fixed', 'passing', 'complete', 'works now', before committing, creating PRs, or moving to next task. Also trigger when expressing satisfaction like 'great', 'perfect', or any positive assessment of work state without fresh evidence.

Extract and summarize YouTube video content via subtitle extraction. Trigger when user shares a YouTube URL (youtube.com or youtu.be), says 'summarize this video', 'watch this', '看看这个视频', or wants to understand video content without watching. Also trigger for video transcript extraction.