데이터 분석, 머신러닝, LLM 앱, 프롬프트 설계 (9천건)
Pre-landing PR review. Analyzes diff against the base branch for SQL safety, LLM trust boundary violations, conditional side effects, and other structural issues. Use when asked to "review this PR", "code review", "pre-landing review", or "check my diff".
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Turn any installed skill (gstack, custom, third-party) into an ATO agent the user can summon into war-rooms. Reads a SKILL.md, extracts the persona, strips runtime boilerplate, and writes an agent file at `.claude/agents/<slug>.md` (project-scoped) or `~/.claude/agents/<slug>.md` (global). Prompts for a model roster (primary + 1-2 alts) so cross-family dispatch in war-rooms produces real disagreement. Companion to `ato-warroom` — that skill summons agents this skill creates. Use when asked "turn this skill into an agent", "register X as a war-room agent", or when scoping a new persona before a war-room runs.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Reads, searches and counts Claude Code session transcripts (the .jsonl files under ~/.claude/projects, subagents too) with the csift command-line tool, and ships hooks that re-inject compacted turns or record pending questions. Use when asked what a past or running session said, did, edited, spent or is doing now; when about to grep, jq or script over a transcript jsonl (a type=="user" filter overcounts the person's messages), count how many messages the person typed, tool calls or tokens, or claim something was never done or decided; to rebuild a lost file or plan from what a session wrote, extract pasted images, restore turns a compaction summary lost, find a subagent's own id when CLAUDE_CODE_SESSION_ID names the parent, wait on or message another session or subagent; or when csift prints "unknown label selector", "show needs an address", "looks like a session/agent id, not a project path" or "0 matches — a DEFINITIVE absence". Not for semantic or embedding search: csift matches regular expressions only.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Map a confirmed platform-neutral website Design System to a local Elementor Site Settings and Widget inheritance contract. Use for a new or rebuilt Elementor site that needs Global Style setup guidance, or an existing site that must preserve its current visual system. Do not redesign the site, implement Widgets, or modify WordPress, Elementor, themes, or remote data.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Schedule a prompt to run later, once or on a repeating cron expression.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
TanStack Table best practices for building headless, type-safe data tables in React with sorting, filtering, pagination, row selection, and column management. Use when building data grids, implementing client-side or server-side table features, defining column structures, managing table state, or optimizing table rendering performance.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
TanStack Router best practices for type-safe routing, file-based routing, data loading, search params, and navigation in React. Use when building React SPAs with complex routing, implementing type-safe search params, setting up route loaders, integrating with TanStack Query, or configuring code splitting and preloading.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Next.js best practices - file conventions, RSC boundaries, data patterns, async APIs, metadata, error handling, route handlers, image/font optimization, bundling
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
TanStack Query (React Query) v5 best practices for data fetching, caching, mutations, and server state management. Use when building data-driven React applications, setting up query configurations, implementing mutations/optimistic updates, configuring caching strategies, integrating with SSR, or fixing v4→v5 migration errors.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Work with the @upstash/search TypeScript/JavaScript SDK, a serverless full-text and semantic search database with built-in reranking. Use when adding search to an app or site, creating a search index, upserting documents with searchable content and filterable metadata, running keyword, semantic, or hybrid search queries, reranking results, filtering with SQL-like or structured filter syntax, paginating with range, fetching or deleting documents, resetting an index, or checking index info. Also use when the user asks for site search, product, document, or knowledge-base search, or a managed search service that needs no cluster to run.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
个性化雅思口语语料生成器。纯 Agent 内闭环（问卷 → 引导对话 → 7 步蒸馏 → 个人画像 → 静态语料网站），零后端依赖，一键安装可用。需要题库自动同步 / RAG 检索可升级到 runnable-export 模式。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
对 DiT（扩散/多模态生成）模型执行 W8A8 动态（data-free）或 W8A8 MXFP8（data-free）量化。覆盖 HunyuanVideo、Wan2.2-T2V/I2V/TI2V、FLUX.1-dev（已迁移）、SD3、Sana、HunyuanDiT、CogViewX 等。YAML apiversion 固定 multimodal_sd_modelslim_v1，统一 MultimodalPipelineInterface 重构路径（不再保留 LegacyMultimodalPipelineInterface）。模板与字段以 msmodelslim/model/hunyuan_video/model_adapter.py（唯一已迁移到重构路径且 lab_practice YAML 已 validated 的 DiT）为参考；其它 DiT 需按各自推理仓的 `parse_args` 调整字段集合。data-free 指不引入外部激活校准数据，spec.dataset（校准集短名，短名如 wan2_2_t2v；enable_dump: false 时不参与 dump）仍必填；精度调优通过下游经验库 L2 §7 + 直接调 `quantization-expert-experience-tuning-rules/scripts/apply_rollback.py`（整层回退）扩展 exclude 列表，闭环由 `quantization-accuracy-tuning-orchestrator` 调度。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
在 W8A8 动态量化后，用 MindIE-SD 冒烟测试 DiT 量化权重是否可加载（可选 skill）。与 `quant-tuning-evaluate`（DiT 扩展节）的区别：本 skill 仅做"能跑通 + 权重可用"的轻量验证（4~16 条 prompt，几分钟出结果），不算分；后者产出视频集交 `quant-tuning-score-dit`（AISBench-VBench 16 维度）评分，是调优回路一环。不引入 vLLM-Ascend、不做精度指标评估（FID/CLIP-score 等）。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Ascend cluster comparison tool. Invoke when user asks to compare two cluster datasets (DB or TEXT), or to generate cluster_analysis_output from raw profiling data via msprof-analyze and then compare.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Analyze Ascend NPU schedule, operator dispatch, operator launch, and Host Bound profiling issues in Ascend profiling data. Use when need to diagnose device Free time, framework/operator dispatch latency, launch latency, PYTORCH_API/CANN_API launch gaps, aclrtSynchronizeStream stalls, task queue behavior, CPU scheduling interference, GC/lock pauses, CPU affinity, or schedule-side optimization actions.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
为 msModelSlim 创建基础 Transformers 模型适配器（Model Adapter）。 包含创建适配器、实现必需接口与注册安装流程。 适用：Decoder-only LLM、理解类 VLM（仅 LLM/text 部分）。 多模态生成（DiT/扩散，含双/多专家）见下方「多模态生成扩展」节，按需加载。 不适用：Encoder-only、非 Transformers 架构。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Analyze Ascend NPU collective communication profiling data with a DB-first workflow. Use when the user provides `cluster_analysis_output/cluster_analysis.db`, rank-level `analysis.db`, rank-level `ascend_pytorch_profiler_{rank_id}.db`, together with `profiler_info.json`, and asks about HCCL or hcom communication cost, collective communication TOP ops, wait time, slow rank/straggler, Notify Wait, bandwidth, retry, relay, SDMA/RDMA/HCCS links, communication matrix, or Ascend communication fault patterns.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
面向 Ascend PyTorch Profiler / msprof DB（如 ascend_pytorch_profiler*.db、msprof_*.db）的 SQL 分析技能。将自然语言问题（算子耗时、通信、下发、调度、schema/table 查询）转为安全可执行 SQL，并按需从官方文档提取表结构详情。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Analyze Ascend NPU computation-side profiling data for single-card runs or a selected rank from multi-card runs. Use this skill when the user asks to diagnose computation bottlenecks, AI Core / AI Vector / AICPU hotspots, dynamic shape overhead, block dim issues, redundant TransData/Transpose/Cast, cross-stream waits, frequency/runtime-state issues, or fusion opportunities.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Compares two Prof datasets (GPU vs NPU / NPU vs NPU) by auto-installing and calling msprof-analyze compare, then parses the result xlsx into HTML report and Chinese xlsx. Invoke when user provides two prof dirs to compare, or has a compare result xlsx to analyze/translate.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
执行模型测评（统一入口）。LLM/VLM 走主流程 scripts/run_evaluation.py + Evaluation YAML；model_family=dit 走 DiT 扩展节（references/dit/evaluate_workflow.md，vbench.py 批量推理产出视频）。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
当用户直接要求进行数据集压缩测评，或在量化调优过程中需要使用 AISBench 评测工具输出数据集精度等场景时使用。 本 Skill 负责：使用 AISBench 数据集压缩代码、用 RBF Kernel Herding 算法生成压缩后的数据子集（coreset），并按用户目标交付两种产物之一： ① 接入量化调优主流程：产出独立的 coreset 数据集配置（不覆盖原始数据），回传「全集 config_name + coreset config_name」给编排层，供 evaluation.yaml 切换； ② 仅数据集压缩测评（不做量化调优）：基于 example 模板生成「全集测试脚本」和「子集测试脚本」两份 shell 脚本。 暂时支持 AIME 2025 和 GPQA，其他数据集需要再结合 aisbench tools目录下 herding_coreset_selector 源码进行分析。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
当用户提供 MindStudio profiler 采集的性能数据（框架 profiler、msprof 命令行）时，对数据完整性、采集状态及关键配置进行校验，确保后续分析工具能正常运行。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use when diagnosing NPU + PyTorch or LLM Serving Host CPU affinity, NUMA locality, cgroup/cpuset constraints, CPU range conflicts, PyTorch/runtime threading, DataLoader, tokenizer, scheduler, vLLM-Ascend, SGLang, TTFT, TPOT, tokens/s, QPS, or multi-rank/multi-worker CPU binding issues.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.