tmdgusya

Use when a user's request is vague, ambiguous, or underspecified. Explores the codebase first, then runs an iterative Q&A loop grounded in the findings until ambiguity is gone. Outputs a clear, well-scoped context brief so the user can plan sharply. Triggers on "I want to...", "I need...", "let's build...", "can you help me...", "we should...", or any request where the full scope isn't immediately clear.

Corrective cleanup of AI-generated code — removes LLM-specific patterns while preserving behavior. Use when the user says "deslop", "slop", "clean AI code", "remove AI patterns", or when you spot LLM-generated code smells after any generation session. For general post-change review and fixes ("clean up", "simplify"), use the simplify skill instead.

Orchestrates multi-day execution of complex tasks through milestones. Each milestone goes through plan-crafting, run-plan (worker-validator), and review-work phases with checkpoint/recovery. Runs autonomously end-to-end — no per-milestone approval gates; escalates to the user only on failures. Triggers when the user says "long run", "start long run", "execute milestones", or "run all milestones".

Decomposes complex, multi-day tasks into optimized milestones (ultraplan). Broad parallel exploration grounds a draft milestone DAG, then independent adversarial critics attack the draft in rounds until no blocking findings remain. Triggers when the user says "plan milestones", "break this into milestones", "ultraplan", or when long-run harness needs milestone generation.

Use when a task's scope is clear and multi-step implementation is needed, before touching code. Triggered after clarification is complete, or when the user explicitly requests plan creation with a clear prompt.

Use after run-plan completes to independently verify the implementation. Reads only the plan document and inspects the codebase from scratch — information-isolated from the execution context. Produces a structured review document with PASS/FAIL verdict. Triggers when the user says "review the work", "verify the implementation", "check if the plan was executed correctly".

Use when you have a written implementation plan to execute. Loads the plan, reviews critically, executes tasks in dependency order, and reports completion. Triggers when the user says "run the plan", "execute the plan", or "let's start implementing".

Use when encountering any bug, test failure, or unexpected behavior. Enforces a strict reproduce-first, root-cause-first, failing-test-first debugging workflow before fixing.

Use when a user's request is vague, ambiguous, or underspecified. Runs recon-first clarification — codebase exploration before questions, then one bundled question round with defensible defaults — and outputs a Goal Contract for the durable /goal runtime. Triggers on "I want to...", "I need...", "let's build...", "can you help me...", "we should...", or any request where the full scope isn't immediately clear.

Primary execution workflow for durable /goal runtime. Use when a Goal Contract is active or when the user asks to execute, continue, verify, or complete a goal.

Review changed code for reuse opportunities, quality issues, and inefficiencies using three parallel review agents, then fix any issues found. Triggers when the user says "agentic-simplify", "clean up the code", "review the changes", or after goal implementation when code quality verification is needed.

Use when encountering any bug, test failure, or unexpected behavior. Enforces a strict reproduce-first, root-cause-first, failing-test-first debugging workflow before fixing.

Interactive idea development through guided Q&A dialogue. This skill helps users clarify and develop their ideas by asking targeted questions, expanding on possibilities, and producing a structured markdown document capturing the essence of their thinking. Triggers: "brainstorm", "idea", "organize ideas", "I want to organize my thoughts", "whatever comes to mind"

Behavioral guardrails to prevent common LLM coding mistakes — enforces surgical changes, assumption verification, and scope discipline before and during implementation. Use when implementing features, modifying code, or when you notice yourself about to make changes without reading the existing code first.

Rob Pike's 5 Rules of Programming — a decision framework that prevents premature optimization and enforces measurement-driven development. Use when the user says "optimize", "slow", "performance", "bottleneck", "speed up", "make faster", "too slow", or any request to improve code speed/efficiency. Also use when you notice yourself about to suggest a performance optimization without measurement data. This is a thinking discipline, not a tooling workflow.

Presentation deck mode. Use when the user wants a presentation, slide deck, or PPTX built from a topic or script — especially when they say "발표", "슬라이드", "프레젠테이션", "발표 자료 만들어줘", "슬라이드 뽑아줘", "presentation", "slide deck", "pptx", "deck". Secures a script, chains into crucible for design tokens (resume protocol), then builds a token-styled self-contained HTML deck. Hooks enforce the gates.

Storyline-driven design mode. Use when the user makes a storyline- or reference-driven design request — design tokens, mood/visual identity, a style guide — especially when they say "디자인", "무드", "스토리", "레퍼런스", "design tokens", "style guide". Prospects real references with a multi-agent panel, smelts them into W3C design tokens, assays fidelity to the story, then requires approval from a 2-designer critic panel before handing off to forge. Hooks enforce the gates.

Strong planning mode. Use when the user requests a non-trivial feature, refactor, or system change — especially when the request is ambiguous, spans multiple files, or says "제대로", "강하게", "빡세게", "thoroughly", "strong plan". Runs deep codebase recon BEFORE asking any question, drafts an executable plan, then requires unanimous approval from a panel of 3+ critic subagents before the plan may be presented. Hooks enforce the critic gate.

Implementation loop. Use when a plan document exists (from forge or blueprint) and implementation should run — including when the user says "implement", "execute the plan", "구현해", "실행해", or approves a plan. Executes tasks through a worker → validator → implementation-critic loop, then a security/QA review gate; any failure loops back to rework. Hooks block turn completion until every task and review passes.

Light planning mode. Use when the user requests a small, clearly-scoped change (roughly ≤3 files, no new interfaces, no ambiguity) that still deserves a written plan before code. For ambiguous or multi-component work, use the forge skill instead. Produces an executable plan with binary acceptance criteria and a self-review, then hands off to hammer.

Use when memento-reflect identifies a reusable pattern worth capturing, or when the user explicitly wants to create a new learned skill from a task execution pattern

Use when a learned skill has low utility score, has been failing, or the user wants to improve an existing learned skill based on failure analysis

Use when the user wants to remove low-value learned skills, clean up unused skills, or reduce the learned skill library size

Use after completing any significant task to self-evaluate what worked or failed, update learned skill metrics, and discover reusable patterns worth capturing as new skills

Use when the user wants to see an overview of their learned skills, check utility scores, or identify skills that need optimization or pruning

Manage tasks on a Mission Board kanban — list, create, claim, update, complete, release, and review tasks via the Mission Board API. Use when the user says /mission or asks about missions/tasks on the board.

Implements Hono API endpoints, database operations, and business logic with Bun/SQLite

Implements CLI tool with command handlers, API client, and Bun compilation to standalone binary

Implements React dashboard components, state management, and UI features

구현 전에 의도 정렬을 절차로 강제한다. 코드베이스를 먼저 탐색하고 모든 모호함을 끄집어내 사용자와 합의할 때까지 반복한 뒤, 정렬된 의도를 handoff 문서(의도 + 확정 범위 + 검증 가능한 완료 조건)로 남긴다. 산출물은 코드가 아니다. 다단계 구현 전, 요청이 모호할 때, 또는 사용자가 "align", "정렬", "의도 맞춰", "문맥 파악해", "명확히 해줘"라고 할 때 사용한다. 스스로 트리거로 쓸 수도 있다 — 요청이 모호해 보이면 찍어서 물어보지 말고 align에 들어간다.

align이 만든 handoff 문서의 "완료" 조건을 ZCode /goal이 판정할 수 있는 객관적 인수조건(실행 명령 + 예상 결과 쌍)으로 변환합니다. 산출물은 인수조건 목록이며 코드는 쓰지 않습니다. handoff가 있을 때, 또는 사용자가 "forge", "인수조건 만들어", "검증 조건 세워"라고 할 때 사용하세요.

검증·완수·조사를 절차로 강제해서 GLM-5.2(ZCode의 모델)가 작업을 끝까지 밀고 가게 만드는 하네스. 강한 에이전트 모델의 작업 습관을 GLM-5.2로 가져온다. 다단계 작업(2개 이상의 순차 목표), 긴 자율 작업, 디버깅/원인 조사, 렌더·실행 산출물(HTML, SVG, 게임, 차트)을 만들 때, 또는 사용자가 "prometheus", "끝까지 해줘", "verify as you go", "목표로 쪼개줘"라고 할 때 사용한다. 증거 없는 "완료"를 허락하지 않는다 — 작업이 끝났다고 선언하기 전에 직접 실행/관측한 사실로 확인하라. 다중 스토리에서는 각 스토리를 검증 가능한 /goal 문장으로 바꿔 사용자에게 제안하고, 사용자가 /goal(또는 /goal replace)로 설정하면 ZCode 런타임이 매 턴마다 완료 여부를 판정한다.

Use when writing git commit messages for non-trivial changes — captures decision context (constraints, rejected alternatives, confidence, directives) as structured git trailers so future agents and developers can query project knowledge via git log --trailer=

Use when the user wants to query Lore git trailers from commit history — surface constraints, rejected alternatives, directives, test gaps, and other decision context. Triggers on "최근 제약조건 확인", "리젝된 대안 보여줘", "check recent constraints", "show directives", or /lore-query.

Use when the user wants to configure Lore commit format in their project or globally — writes Lore rules to AGENTS.md, CLAUDE.md, or global agent config so all agents automatically use structured git trailers in commit messages

Automates browser interactions for web browsing, scraping, form filling, screenshots, and UI interaction. Use when the user needs to visit a webpage, check a live site, scrape content, fill forms, take screenshots, or interact with a web UI.

Automates browser interactions for web browsing, scraping, form filling, screenshots, and UI interaction. Use when the user needs to visit a webpage, check a live site, scrape content, fill forms, take screenshots, or interact with a web UI.

Scrape posts from Threads (threads.com) profiles. Handles login, infinite scroll, and post extraction. Use when the user wants to fetch posts from a Threads profile.

Create a comprehensive Test Guide using the Oracle Test-First Architect methodology before implementation

Create a verification file (shell script or markdown checklist) for use with the more-loop iterative development script

Create a prompt.md spec file for use with the more-loop iterative development script

Creates detailed implementation plans after requirements are clarified. Use when user says "plan this", "how should I implement", or after requirements-clarifier skill has gathered context. Breaks down work into steps with dependencies and complexity estimates.

This skill should be used when the user asks to "scan the codebase", "analyze multiple patterns", "parallel agent workflow", or "multi-agent research". Spawns multiple agents in parallel for comprehensive codebase analysis.

Activates during planning stage to clarify user requirements, constraints, and preferences. Use when user requests are vague, ambiguous, or need deeper understanding before implementation. Helps prevent rework by ensuring shared understanding.

Set up and use the hermes-provider-switcher plugin to run Claude Code with alternative LLM providers (GLM, Kimi, MiniMax).

Set up API keys for hermes-provider-switcher plugin (GLM, Kimi, MiniMax). Supports fresh setup or migration from shell rc files.

Use when the user asks to draw diagrams, create visuals, sketch wireframes, or interact with tldraw desktop canvas via its local HTTP API on port 7236. Triggers on "draw", "diagram", "sketch", "wireframe", "flowchart", "tldraw", "canvas".