개발 스킬

Coordinate safely with a peer interactive session sharing the same git worktree — check first whether a separate worktree removes the problem, then declare path and branch scope before the first edit, survive a contended index lock, and review the whole branch before opening a PR. Activate when starting work in a repository that may already be occupied, when a git command fails with `index.lock: File exists`, when `guard:snapshot` refuses because a snapshot already exists, or when a commit turns out to contain a file this session never edited. When one session leads and the other supports, divide the labour by shape and re-derive every reported number against a named revision before citing it. Distinct from subagent concurrency: a peer session cannot be bracketed, because no baseline exists from before its edits.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Create parameterized Quarto or R Markdown reports that can be rendered with different inputs to generate multiple variations. Covers parameter definitions, programmatic rendering, and batch generation. Use when generating the same report for different departments, regions, or time periods; creating client-specific reports from a single template; building dashboards that filter to specific subsets; or automating recurring reports with varying inputs.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Analyze pre-trained generative diffusion models (Stable Diffusion, DALL-E, Flux) by computing quality metrics (FID, IS, CLIP score, precision/recall), inspecting noise schedules, extracting and visualizing attention maps, and probing latent spaces. Use when evaluating a pre-trained generative diffusion model's output quality, comparing noise schedule variants, analyzing cross-attention patterns for text-conditioned generation, interpolating between latent codes, or detecting out-of-distribution inputs.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Four-stage alchemical code transmutation — nigredo (decomposition), albedo (purification), citrinitas (illumination), rubedo (synthesis) — with meditate and heal checkpoints between stages. Transforms tangled or legacy code into optimized, well-structured output through systematic material analysis. Use when transforming legacy code into modern equivalents, refactoring deeply tangled modules where incremental fixes keep failing, converting a codebase between paradigms, or when simpler refactoring approaches have stalled and a full-cycle transformation is needed.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Analyze an arbitrary codebase to identify functions, APIs, and data sources suitable for exposure as MCP tools, producing a tool specification document. Use when planning an MCP server for an existing project, auditing a codebase before wrapping it as an AI-accessible tool surface, comparing what a codebase can do versus what is already exposed via MCP, or generating a tool spec to hand off to scaffold-mcp-server.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Build a plugin or adapter for a CLI tool using the abstract base class pattern. Covers defining the contract (static fields, required methods), choosing an installation strategy (symlink, copy, append-to-file), implementing detection, install/uninstall with idempotency, listing, auditing, and registering the plugin. Use when adding support for a new framework to a CLI installer, building a plugin system for any multi-target tool, or extending an existing adapter architecture.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Build a custom MCP (Model Context Protocol) server that exposes domain-specific tools to AI assistants. Covers server implementation in Node.js or R, tool definitions, the resources and prompts primitives, transport configuration, and testing with Claude Code. Use when you need to expose custom functionality beyond what mcptools provides, when building specialized domain-specific AI integrations, or when wrapping existing APIs or services as MCP tools.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Address unwanted dog behaviors through desensitization, counter-conditioning, and management. Covers reactivity (to dogs, people, sounds), separation anxiety, resource guarding, excessive barking, and leash pulling using force-free methods with systematic threshold management. Use when a dog shows reactivity toward other dogs, people, or stimuli; when separation anxiety causes destructive behavior or vocalization; when resource guarding is present; or when behaviors like excessive barking or leash pulling interfere with daily life — after basic obedience is established.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Guide for adding new Electron APIs to Wave Terminal. Use when implementing new frontend-to-electron communications via preload/IPC.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Guide for adding new wsh commands to Wave Terminal. Use when implementing new CLI commands, adding command-line functionality, or extending the wsh command interface.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

List configured models, guide provider setup, verify local Kimi companion readiness, and manage the PreToolUse hook plus optional review gate. Use when asked about available models, subscription/API-key setup, default selection, or explicitly requested to install, check, enable, disable, or uninstall the integration. Codex and Claude Code share one ~/.kimi-code/config.toml but each own a host-scoped block, so $kimi-setup here does not disturb Claude Code's /kimi:setup (and vice-versa).

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use when checking Agent Brain inside a real agent runtime or adapter before trusting harness usability.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use when project vocabulary is fuzzy, overloaded, disputed, or needed before naming docs, schemas, commands, skills, code, plans, or memory.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use when agent-produced work needs a safety, evidence, and reliability check before handoff, merge, or trust.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use when a command, adapter, or runtime smoke depends on proving an agent runtime's capabilities before trusting command routing, writes, shell access, or validation claims.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

一键发布 agent skill 到 GitHub，自动验证 SKILL.md、生成或检查更吸引人的 README、区分 skill name 与 GitHub repo name、创建或更新仓库、推送并通过 npx skills 真实安装验证。当用户说"发布这个skill"、"publish skill"、"把skill发到GitHub"、"分享这个skill"、"/publish-skill"时触发。支持发布当前目录或指定路径的 skill。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Policy for the cross-model council: when to hand off to an installed sibling plugin's own slash command (`/codex:*`) versus dispatching our `delegation-runner` (backend: codex|agy|…), which backend to pick, what permission posture to use, and the fan-out/cost discipline that applies regardless of backend — plus the council layer built on that substrate: which play to run, how a cast gets picked from the project roster, budget guardrails, and where project policy/commission config lives. Consult before delegating out to a peer agentic CLI, before choosing between a sibling plugin's command and our runner, before choosing a permission posture, before fanning out more than one external delegation concurrently, or before running a council play (`/…:council-run`).

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Scaffold a new Android app with SDLC and Play-safe defaults — Kotlin/Compose Gradle project with version catalog, convention plugins, module layout, a targetSdk that meets the Play mandate, R8 on release, and git-ignored signing so it passes the pre-flight checker from day one. Use when starting a new Android project.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Configure SDLC team formation and SDLC method for this project. Asks the four-option SDLC method question (solo / single-team / programme / assured), then recommends and installs team plugins based on project type, language, tech stack, and support needs.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Set up Android app signing safely — generate an upload keystore, enroll in Play App Signing, and wire a Gradle release signingConfig that reads secrets from a git-ignored keystore.properties / CI (never committed). Use when configuring signing for a Play release.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

AI-First SDLC compliance rules and standards. Use when checking code quality, validating architecture, enforcing zero technical debt, or making any compliance decision.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Parallel map-reduce bulk ingest for large source sets. Map phase dispatches read-only knowledge-extractor agents (≤N-wide, default Haiku) emitting per-source JSON extracts; Python routes extracts by target library file (fuzzy-merging new topics, pre-allocating new files, flagging oversized topics); reduce phase dispatches one agent-knowledge-updater per file (parallel, one writer per file); single shelf-index rebuild + one log.md entry. Resumable. Supersedes kb-ingest-batch.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Parse all `# implements:` annotations in source code and emit `library/_code-index.md` as a shelf-index. Also extracts cross-module import edges and writes `library/_dependency-edges.md`. Idempotent — re-running produces byte-identical output if no annotations or imports changed.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Commission a project to an SDLC option bundle (solo / single-team / programme / assured). Asks structured questions, recommends an option × level, installs the bundle, and records the decision in .sdlc/team-config.json.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.