Use when the user asks to "set up conversion values so tROAS optimizes profit not orders", "map margin onto my purchase value", "build value rules for lead / phone / signup conversions", or "stop bidding to revenue when I care about profit"; defines and QAs the conversion VALUE model — per-conversion values, margin/net-value adjustment, static-vs-dynamic value rules, proxy values for non-revenue actions, and a value-vs-count sanity check — as a value-model spec plus a pre-launch value QA sheet. Not for whether the tag fires or UTMs are clean — use conversion-signal-qa; not for cross-platform double-count de-dup — use attribution-reconciler; not for scoring R1/R2 — that is a scored veto in ad-account-auditor. 付费广告转化价值建模/利润出价/价值规则QA

分章节创作引人入胜的中文小说。支持各种题材（悬疑/言情/奇幻/科幻/历史等），支持10-50章长篇创作，每章3000-5000字，结尾设置悬念钩子。强调深度润色去除AI痕迹，确保文字自然流畅。 当用户要求：写小说、创作故事、分章节写作、连续剧情、章节悬念、长篇小说时使用。

Detect IBD-style Distribution Days for QQQ/SPY (close down at least 0.2% on higher volume), track 25-session expiration and 5% invalidation, count d5/d15/d25 clusters, classify market risk (NORMAL/CAUTION/HIGH/SEVERE), and emit TQQQ/QQQ exposure recommendations. Use after market close, before TQQQ exposure changes, or as input to FTD/market-state frameworks. Does not execute trades.

Tally Claude Code session token usage and cost estimates from the raw JSONL conversation log. Trigger when the user asks about session cost, token usage, API spend, cache hit rate, input/output tokens, or wants a breakdown of how much a Claude Code session has cost. Also trigger for "how much have we spent", "show me token usage", "session summary", "cost so far", or any request to analyse or display per-turn metrics from the current or a past session. Do NOT auto-dispatch compare mode (--compare / --compare-prep / --compare-run / --count-tokens-only) from natural-language phrases. The skill body uses $ARGUMENTS[0] as the dispatch key — if the first positional argument is not literally "compare", "compare-prep", "compare-run", or "count-tokens", route to the default single-session report.

钉钉在线文字文档（adoc）本体及其内容：查找、创建、读取、文档信息、编辑、块、评论、附件与媒体、白板卡片容器插入删除、导入、导出(docx/markdown/pdf)、版本、模板、协作者权限、分享及Markdown/JSONML写入。不包括：白板/画布内的 OpenNodes 图形、文本、分组、连线、Vector 和布局读写（归 dingtalk-misc 的 Whiteboard）、文档空间与钉盘的文件管理（归 dingtalk-drive，doc 同名原子命令已弃用）、知识库空间与节点管理（归 dingtalk-wiki）、原生 .md 文件读写（归 dingtalk-misc）、电子表格 axls（归 dingtalk-misc）、AI 表格 able（归 dingtalk-aitable）。命令前缀：dws doc。

Attach an image or video to a GitHub pull request or issue with `gh --attach`, as a comment or in the description. Use for "add a screenshot to the PR", "put this image in the issue", "show before/after in the description".

Stateful adoption coach for LeGreffier: inspects local and remote state, classifies the current adoption stage, and suggests the next best action. Use when getting started with LeGreffier, after installing the plugin or running moltnet agents init, when asked "what should I do next", "how do I use legreffier", "set up diary", "connect team diary", or "onboarding".

Guardrails for writing, modifying, reviewing, or refactoring code. Use whenever a task involves code to clarify ambiguous requirements before implementation, avoid over-engineering and speculative abstractions, and keep changes minimal and scoped.

当用户明确要求“迁移 LaTeX 模板”“把旧项目接入 ChineseResearchLaTeX”“把旧标书/论文/毕业论文/简历套进当前模板”“把 Word/PDF/Markdown/零散 tex 整理进现有项目”，或直接提到 `transfer-old-latex-to-new` 时使用。旧别名 `migrating-latex-templates` 可兼容理解。该 skill 只负责把正文内容迁移到当前仓库现有模板的内容层；绝不能修改 `packages/` 内公共包源码、也绝不能修改 `projects/` 内模板样式或入口骨架，只能写入目标项目允许承载正文的内容文件。

当用户给出期刊/杂志名并希望了解投稿要求、投稿形式/格式清单、期刊官网信息、社区评价、审稿速度、费用、文章类型、格式细节或“这个 journal 怎么投/是否靠谱/投稿指南调研”时必须使用。联网核验期刊官网与社区评测，默认把全部中间文件隔离到当前工作目录的 `.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/paper-know-journal/`，最终只交付 `KnowJournal-{杂志名}.md`。⚠️ 不适用：用户要根据 manuscript 选择投稿期刊（用 paper-select-journal）、写论文正文、下载全文 PDF，或只问一个无需成稿的简单事实。

当用户明确要求"写/改研究基础""研究基础+工作条件+风险应对编排"时使用。为 NSFC 正文"（三）研究基础"写作/重构，并同步编排"工作条件"和"研究风险应对"，用证据链证明项目可行、资源条件对位研究内容、风险预案可执行。

当用户明确要求"写/改研究内容""研究内容+创新+年度计划编排"时使用。为 NSFC 正文"（二）研究内容"写作/重构，并同步编排"特色与创新"和"三年年度研究计划"，输出可直接落到 LaTeX 模板的三个 extraTex 文件。

当用户明确要求"标书QC/质量控制/润色前质检/引用真伪核查/篇幅与结构检查"时使用。对 NSFC 标书进行只读质量控制：并行多线程独立检查文风生硬、引用假引/错引风险、篇幅与章节分布、逻辑清晰度等，最终输出标准化 QC 报告；中间文件默认归档到 `project_root/.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/nsfc-qc/{yyyy-mm-dd-hh-mm}/`，并兼容 legacy `.nsfc-qc/` 读取/清理。

当用户要求写作、重构、审查或润色 NSFC/科研基金申请书的立项依据、研究意义、国内外现状、科学问题或科学假设时使用。以语义论证和可核验性为核心，支持任意文件名、标题命令和 LaTeX 结构；默认全自动完成诊断、成稿、校正、备份和写入，不向用户发起中途确认或审核问题。

当用户明确要求"写/润色 NSFC 标书摘要""生成中文摘要和英文摘要""把中文摘要翻译成英文摘要"时使用。输出中文、英文两个版本（英文必须是中文的忠实翻译版），同时输出标题建议（1个推荐标题+5个候选标题及理由）。中文摘要默认≤400字符，英文摘要默认≤4000字符。输出方式：将结果写入工作目录下的 `NSFC-ABSTRACTS.md`。⚠️ 不适用：用户只想翻译一段与标书无关的通用文本（应直接翻译）；用户只想写立项依据/研究内容/研究基础正文（应使用对应 nsfc 系列 skill）。

Visual ChangeNet for binary image classification and segmentation in AOI defect detection. Use when training, evaluating, exporting, or running inference for PCB defect detection or visual inspection, comparing image pairs for PASS/NO_PASS classification, or producing change-segmentation masks. Trigger phrases include "train Visual ChangeNet", "ChangeNet classify", "ChangeNet segment", "AOI defect detection", "PCB inspection model".

Real-time stereo depth estimation using FastFoundationStereo (FFS), the distilled bp2 commercial variant of FoundationStereo. Predicts disparity maps from stereo image pairs with ~10× lower latency than full FoundationStereo. Use when training, evaluating, exporting, or running inference for a TAO FastFoundationStereo (FFS) model. Trigger phrases include "train fast stereo", "real-time stereo disparity", "FastFoundationStereo", "distilled stereo depth".

Fine-tune any HuggingFace CV / VLM / LLM model on local NVIDIA GPUs inside an NGC PyTorch container when no dedicated TAO model skill matches. Use when the user wants to fine-tune a HuggingFace model (full or LoRA), train a vision / VLM / LLM model end-to-end, generate a reproducible HF training pipeline, smoke-test a HuggingFace model locally before scale-up, push a fine-tuned model to the HF Hub with a model card, or emit a self-contained rerun skill for an existing HuggingFace finetune. Supports image classification, object detection, semantic / instance / panoptic segmentation, depth estimation, image-text-to-text VLM (SFT / LoRA), and LLM SFT / DPO / GRPO. Six-step workflow: inspect and qualify, hardware and NGC image, research, generate and smoke, train + eval + infer, push and emit rerun skill. Do not use for any Hugging Face model ID claimed by a dedicated `skills/models/*` skill; the model skill and its declared execution environment take precedence.

Use when planning, debugging, tuning, evaluating, exporting, or deploying public Nemotron `embed`/`rerank` retrieval recipes.

Plan, configure, and chain repo-native Nemotron customization steps into single-step or multi-step pipelines: curation, translation, SFT/PEFT (AutoModel or Megatron-Bridge), pretraining/CPT, RL alignment (DPO/RLVR/GRPO/RLHF), BYOB/MCQ benchmarks, checkpoint conversion, ModelOpt optimization, env profiles, and evaluation of trained checkpoints or existing/hosted endpoints. Use when a request names a Nemotron step or workflow, or asks to clean, translate, train, fine-tune, align, convert, optimize, evaluate, or compose these into a pipeline. Do NOT use for frontend/dashboard/visualization work, generic ML advice, billing/access, or non-Nemotron coding tasks.