9,131 件 ・ 関連度順
概要と使いどころ
Apply a refined Scandinavian visual system with a neutral black-and-white foundation, restrained sans-serif typography, purposeful product imagery, generous spacing, clear page chapters, and minimal decoration. Use when the user invokes /scandinavian-design or requests a Nordic, monochrome, calm, spacious, restrained, or minimalist interface.
日本語の概要は準備中です。原文の説明を表示しています。
Reference when designing autonomous agent systems or deciding agent freedom/safety/context management/collaboration infrastructure. 10 design principles distilled from Karpathy autoresearch (autonomous ML experiment loop) + AgentHub (agent-native Git DAG). Triggered by "autoresearch", "agenthub", "ratchet" ("래칫"), "context budget", "dumb platform", "program.md", "overnight", "autonomous experiment" ("자율 실험"), "branchless", "single-file constraint" ("단일 파일 제약"), "fixed time budget" ("고정 시간 예산") keywords.
日本語の概要は準備中です。原文の説明を表示しています。
Test-Driven Development workflow for Software Metrics Machine. Covers test frameworks (Vitest 4.x for CLI/REST/core/utils, Jest 30.x for webapp), test patterns (describe/it, builder pattern, mocks via vi.fn), test commands, coverage requirements, and the Red-Green-Refactor cycle. USE FOR: write tests, run tests, test this, add tests, testing, coverage, TDD, test-driven development, vitest, jest, test pattern, test convention. DO NOT USE FOR: linting (use lint skill), building (use build commands directly), debugging production issues.
日本語の概要は準備中です。原文の説明を表示しています。
Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.
日本語の概要は準備中です。原文の説明を表示しています。
Use the Xero MCP server — troubleshoot its self-refreshing OAuth2 token file, maintain the local best-of-breed server build, and pick up other operational notes for working with Xero MCP tools. Use when Xero MCP tools fail with authentication errors, before starting any Xero workflow, when adding or updating upstream xero-mcp-server PRs, or for general guidance on Xero MCP usage.
日本語の概要は準備中です。原文の説明を表示しています。
Applies local reliability, viewport, tab-safety, stale-ref, widget, and browser-boundary lessons to every agent-browser automation. Use alongside the upstream agent-browser skill whenever navigating, clicking, filling, testing, extracting, taking screenshots, or debugging any website with agent-browser.
日本語の概要は準備中です。原文の説明を表示しています。
Root-causes one specific mistake the agent just made, such as not committing despite instructions to, by checking it against the loaded AGENTS.md, CLAUDE.md, skills, and Beads memories, delegating the investigation to a separate agent where the harness supports one; proposes concrete context improvements, and lets the user choose which to apply and how to remediate. Use when the user invokes `/learn` or `$learn`, with or without a description of the mistake, including `/learn /qs`-style shorthand naming the corrective command. For a broad end-of-session documentation sweep, use `documentation-updates` (`/reflect`) instead.
日本語の概要は準備中です。原文の説明を表示しています。
Deletes files and directories through a wrapper that avoids the usual per- command permission prompt. Use only when content must genuinely be deleted, such as clearing build artifacts or dropping obsolete files during a refactor. Moves, renames, and relocations that preserve content use mv instead.
日本語の概要は準備中です。原文の説明を表示しています。
Enforces Beads policy for a repository maintained initially by one owner, granting issue-management and commit authority while requiring explicit permission for Git pushes (other than to the user's own fork) and Dolt sync or push operations. Use when the user asks to enroll, repair, or operate a solo-maintainer Beads workspace, and when working with Beads in a repository that may carry a beads-solo enrollment, including when bd refuses a schema migration or mentions a remote.
日本語の概要は準備中です。原文の説明を表示しています。
Evaluate repeated code and consolidate duplication when a shared abstraction improves correctness or maintenance. Use after introducing similar logic or when repeated changes reveal a missing source of truth.
日本語の概要は準備中です。原文の説明を表示しています。
Create subagent definitions for Claude Code and OpenCode that delegate to skills. Use when creating new subagents or refactoring existing ones to follow the delegation pattern.
日本語の概要は準備中です。原文の説明を表示しています。
Plan dependency-aware atomic commits for changes that span multiple logical concerns. Use when a refactor, feature, or API change would benefit from reviewable, independently meaningful commit boundaries. To triage an already-dirty working tree and propose commits for existing uncommitted changes, use `planning-commits` instead.
日本語の概要は準備中です。原文の説明を表示しています。
Create Claude Code slash commands, OpenCode command files, Pi prompt templates, and Codex custom prompts that delegate to the right subagent or skill. Use when creating new commands, porting commands across agent platforms, or refactoring existing commands to follow platform conventions.
日本語の概要は準備中です。原文の説明を表示しています。
Corrects output that refers to entities by bare identifier, restating it with a human-readable label beside every ID, and keeps that rule for the rest of the conversation. Use when the user invokes `/ids` or `$ids`, or complains that bead IDs, PR numbers, commit SHAs, or ticket keys were given without titles.
日本語の概要は準備中です。原文の説明を表示しています。
Cross-reference diffs to update README, ARCHITECTURE, CHANGELOG.
日本語の概要は準備中です。原文の説明を表示しています。
Test-Driven Development for backend code. Red-Green-Refactor cycle with practical patterns.
日本語の概要は準備中です。原文の説明を表示しています。
分布式并行策略选型与实测（USP / CP 通信掩盖 / CFG / TP/RSP/PP 概览；含拓扑相关选型 （按实测拓扑分域条件化：域内 bulk vs 跨域 head-parallel 翻转）与 AlltoAllV 缺陷绕过）。在 model-auto-optimization 中承担 S3：优先 USP、结合拓扑带宽差异选 CP，少量 step + 多 rank 验证特性开启与掩盖（产物：并行方案 + 多 rank 证据）。当用户需要多卡并行 策略选择、**序列并行形态抉择**（纯 Ulysses vs 复合 AllGather-KV×Ulysses：按 GQA / 跨域带宽 / 形态 plumbing 条件化定胜负）、**并行 × 稀疏叠加**（seam 契约：先汇聚后稀疏、 窗口偏移、块对齐、per-head 掩码；含「稀疏看似生效实则未生效」判定）、通信掩盖调优 （含**掩盖率上限**：1-1/n 何时成立、c/f 决定的真实上限、没生效的排查）、**并行方案 差异归因**（阶段 Δ 分解 / 集合通信按 communicator 归属 / 4→8 卡线性度），或排查多卡 跑不动 / 通信暴露大 / 换卡组 / 端口 bind / HCCL 带宽验证问题时使用；**并收编原并行作用域诊断**： 改了 SP/CP/Ulysses/AllGather-KV 后**不报错但结果没变/性能没变**、或小规模能跑大规模崩 （如 Ascend EE1003 coreDim 超限）时，用本技能证明"改动到底有没有生效"（判别量逐层收窄 + 两侧对照 + 日志≠生效，见 `references/scope-effectiveness-check.md`）； 即使用户只说 "多卡跑不动""通信暴露大""为什么没达到 6/7 的掩盖率""CP 和 USP 该选哪个""CP 叠稀疏 怎么不生效"而未说并行，也应触发。特性档位/接口事实见 `docs/zh/features/parallelism.md` /`usp.md`（仓内真源），框架侧开启见 framework-integration； 本技能承载选型决策、monkey-patch 掩盖与多卡诊断实测。 由 dev-workflow 多卡场景触发，亦由 model-auto-optimization 的 S3 阶段触发。
日本語の概要は準備中です。原文の説明を表示しています。
**精度验收标准**：凡是"改动不应改变结果"的场合（等价替换、算子/子模块融合、并行切分、 编译与图下发、拷贝消减），都按本标准判"合格 / 不合格"——等价分层（L1 逐位 / L2 数值门 / L3 有损门）+ 三级验收序（① 同配置重跑逐位 → ② 跨配置数值门 + 产物 md5 不变 → ③ 质量门）+ 判据不达标时的排障入口。当用户说"这个算子能不能换个写法/换个 kernel/等价实现/ 无损替换""替换后结果会不会变""怎么证明逐位一致""结果不对""花屏""尾部塌了" "CPU 跑对 NPU 跑不对"，或发现某个 hotspot 占了大头（例如某类算子在阶段里占 90% 以上）想动手时， 都应触发；即使用户只说"这样改有没有把结果改坏""能不能判它是无损的"也应触发。 覆盖：等价分层与判据、满足本标准的实现约定（per-shape 对拍写进实现，把索引/相位/顺序类重写 错误在首次调用抓住）、per-shape 条件性等价（同一改动在不同形状下结论可能不同）、 判据不达标 → `references/silent-failure-localization.md` 排障入口、否决案例的形态学 （换写法未换 kernel / 差异极小仍非逐位 / 等价但 OOM）、以及收益口径的归属（性能数字一律交 `perf-gate`）。 **近义分流**：选量化档 / 特性选档（要不要开量化、开哪一档）→ `dit-perf-opt`； 并行选型（USP / CP / TP 怎么切）→ `dit-parallel-opt`；本技能只判"结果是否被改变"， 不负责选档与选型。
日本語の概要は準備中です。原文の説明を表示しています。
DiT 计算模块（L3）：把**已定位的 DiT 计算瓶颈**落成特性级选档与实施—— 量化档（W8A16 / W4A16 / W8A8 系列 / W4A4 / MXFP8 / FA 量化）、稀疏（rf_v2 / ada_bsa）、 缓存（DiTCache / AttentionCache / 时间步优化）、编译启用（MindieSDBackend / Pattern 融合 / ACLGraph） 的**开不开、开哪一档、怎么开、怎么复验**；依据是 `docs/zh/features/*`（特性真源）+ framework-integration/references/framework-support-matrix.md（支持状态）。 即使用户只说"这个模型怎么加速""量化/稀疏/Cache 怎么选怎么开""要不要开量化、开哪一档""这个档位开了有没有效果" 而未提 profiling，也应触发。 **入口条件**：瓶颈点已明确（用户带一句实测锚点，或编排层交付标签）时由域入口 `performance-optimization` 按标签分发到本技能；**瓶颈未明（"怎么加速 / 跑通 / 采 profile"）先走 `model-auto-optimization` 定位**，不在本技能内做占比分析。 near-miss：多卡并行形态 / 通信掩盖 / TP·offload 选型 → `dit-parallel-opt`；VAE 解码段与 host 固定开销 → 各自模块（VAE / host）；单算子实现级实测选型（mindie_bench）→ `benchmark-dev`； 需要新增 pattern / 算子才能落地本档 → `pattern-dev` / `operator-dev`；框架侧开关与使能验证 → `framework-integration`；量化器位级契约与精度对齐（编码公式 / 舍入 / scale 粒度）→ `quantization-dev`；精度验收判据 → `accuracy-gate`；数字入库口径 → `perf-gate`。
日本語の概要は準備中です。原文の説明を表示しています。
host / 辅助段优化模块（L3）：把**非 DiT、非解码段**的固定开销落成可复验的优化—— ① 交付与搬运（mp4/图片编码、worker→API 通路、落盘、异步化）； ② 装载与预热（权重加载、编译与图下发预热、镜像/容器预热）。 当用户说“mp4 编码/传输/落盘慢”“交付这段占了不少时间”“模型加载慢”“首启编译太久” “固定开销占比高”“每个请求都要重新编码/重新写盘”时使用。 **入口条件**：占比已达门限（瓶颈标签 `非DiT-host段`；标签枚举、门限与测点口径的单一真源在 model-auto-optimization/references/bottleneck-labels.md，本技能只消费不定义）。 near-miss（看似相关但不属本技能）： - **并发 / 吞吐 / 批处理调度 / 队列深度与限流** → **不属本技能**（会改变所有性能口径与 SLO 语义， 未来独立立项；本技能只在**不改变请求并发模型**的前提下压固定开销）； - VAE/TAE 解码段（计算或通信）→ `vae-opt`；DiT 计算 → `dit-perf-opt`； 多卡并行 / 通信掩盖 → `dit-parallel-opt`； - **装库 / 下载权重 / 构建镜像 / 创建容器** → `env-install`（本技能只优化装载与预热**过程本身**）； - 编译与图下发的**机制本体**（pattern / Inductor / NPUGraph / 图 pool / 缓存条目）→ `pattern-dev` / `aclgraph-dev`（本技能不另做一套缓存）； - 产物一致性与精度判据 → `accuracy-gate`；收益口径 → `perf-gate`； - 瓶颈未定位 / 没有占比锚点 → `model-auto-optimization`。
日本語の概要は準備中です。原文の説明を表示しています。
Comprehensive reference for inter-process communication mechanisms -- every way processes can talk on a computer. Covers sockets (TCP, UDP, Unix domain), WebSockets, SSE, pipes (named/anonymous), shared memory (mmap, shm_open), message queues, signals, D-Bus, XPC (macOS), gRPC, REST, stdin/stdout, file-based coordination, and clipboard. Performance benchmarks, platform availability, and code examples for each. Special focus on which IPC works best for AI agent coordination. Activate on: "IPC", "inter-process communication", "process communication", "how do I talk between processes", "Unix socket vs TCP", "shared memory", "named pipe", "WebSocket vs SSE", "gRPC vs REST", "agent coordination IPC", "XPC service", "message passing", "stdout pipe", "D-Bus", "mmap". NOT for: distributed systems design (use distributed-systems), network protocol design (use networking), message queue infrastructure like Kafka (use data-pipeline-engineer).
日本語の概要は準備中です。原文の説明を表示しています。
Patterns for vibe coding that produce production-quality output instead of garbage. Covers prompt engineering for AI coding, verification without reading every line, accept/reject/refine decision framework, architectural guardrails, context management (CLAUDE.md, .cursor/rules), TDD-driven vibe coding, screenshot-driven development, diff review discipline, and the anti-patterns that destroy quality. Activate on: "vibe coding", "AI coding quality", "Claude Code workflow", "AI-assisted development", "how to use AI for coding", "coding with AI", "agentic engineering workflow", "AI code quality", "vibe code without garbage", "production quality AI code", "CLAUDE.md setup", "cursorrules setup", "pre-commit hooks for AI", "TDD with AI", "test-driven vibe coding". NOT for: multi-person collaborative coding sessions (use cooperative-vibe-coding), building AI/LLM applications (use ai-engineer), prompt engineering for non-coding LLM use cases (use prompt-engineer).
日本語の概要は準備中です。原文の説明を表示しています。
Build and extend pd-console — Port Daddy's GPU-native macOS operator console (GPUI 0.2.x, Zed's Rust UI). Covers the render-agnostic Block/Pane(Surface) contract, the two-thread reqwest↔smol refresh pipeline, Taffy flexbox layout, uniform_list virtual scroll, focus + keyboard nav, the OKLCH theme and ICS maritime flag badges, GPUI's missing text-input, and the real feature-gated cargo/CI gate. Use when adding panes, visual polish, or debugging GPUI rendering/layout/focus in core/pd-console. NOT for the TypeScript daemon, generic Rust toolchain/borrow-checker help (use rust-with-claude-code), or non-pd GPUI apps with a different theme/architecture.
日本語の概要は準備中です。原文の説明を表示しています。
Decision-theoretic framework for agents embedded within the environments they model and act upon
日本語の概要は準備中です。原文の説明を表示しています。