飞书审批：查询和处理审批待办/已办/实例，搜索可发起审批定义、查看定义详情并发起原生审批实例。当用户要处理审批任务、查看审批实例、搜索或发起审批时使用。审批待办不是飞书任务；非审批类待办走 lark-task。不负责创建审批定义；三方审批定义不走原生提单。

Use when asked to build, preview, or verify the Flet documentation website, including checking for broken links, broken images, and unresolved reST cross-references.

构建 GUI 自动化 Agent、让 Agent 像人一样看屏幕并用鼠标键盘操作桌面或浏览器软件时使用。 覆盖感知-思考-行动循环、动作空间（GUI/bash/文件编辑）、视觉定位 Grounding 三条路线 （结构化元素索引、Set-of-Mark、纯坐标预测）、分辨率坐标缩放、关键帧观察接口、世界模型与移动端生态壁垒。 触发词：Computer Use、GUI 自动化、截图、Grounding、browser-use、Set-of-Mark、视觉定位、坐标缩放、OSWorld。

搭建 Coding Agent 或设计 Agent 护栏时使用——当你要构建能自主读写代码、跑测试、自我纠错的编码 Agent，或为已有 Agent 系统补约束/验证/纠正机制、设计执行边界与验收基线、判断某类任务是否适合交给 Agent、优化 Harness 提示词与工具集时使用。覆盖 Harness 四组件、任务四象限、四条设计原则与业界实践。

把生产 bad case / 失败轨迹转成训练数据时使用——从失败归因、轨迹前缀截取、DPO 偏好对构造、LoRA 训练到边界集与保留集双集验证的完整链路；也用于判断某类问题该走 SFT、DPO、RL 还是先修 Harness/工具/tokenizer，以及如何从运行轨迹中提取带证据的结构化学习信号。触发词：bad case、失败归因、DPO、偏好对、轨迹前缀、拒绝采样、过度矫正、边界集、保留集、过早结束、reward hacking、LoRA 微调、训练评估隔离。

Long-term memory for the agent via Hindsight. Use to recall relevant past context before answering, retain durable facts as you learn them, and reflect over accumulated memory for the "why" behind a decision. Load whenever continuity across sessions matters — the user refers to earlier work, states a lasting preference, or asks a question that prior context could answer.

Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.

Write system prompts, tool docs, and agent definitions. Project tag conventions + RFC 2119 keywords + dense compression. Use when authoring or editing any prompt the model reads.

Design a coherent assessment-and-practice lesson using zone-of-proximal-development principles: independent diagnosis, targeted review, supported practice, fresh independent checks, and an evidence-based next step. Use for 习题课（最近发展区）, 最近发展区, 测验习题课, diagnostic review lessons, or guided-to-independent practice. Can also run one practice cycle in chat; do not turn a request for a single exercise or theory explanation into a full lesson.

The design law of a slide page — canvas geometry, the text height table, the type scale, contrast pairs, spacing rhythm, and which element type carries which content. Load it when patching slide elements one at a time, when a fix is about colour or contrast, when text has to grow or shrink inside a box, when a page feels crowded or empty, or when replacement content needs real rich-text structure. It is the standard `page-clone` and `pro-editing` edit against, not a procedure of its own — those two decide which pages to touch and in what order, this one decides what a good page looks like when you put it back.

Infuse learning strategies and metacognition into a concept-centered OpenMAIC classroom as a parallel goal. Use when the user wants students to learn how to learn while studying a subject: retrieval practice, self-explanation, prediction before feedback, monitoring understanding, deliberate review, or productive failure. Do not use when learning science or study skills should be the standalone subject of the lesson.

Use the native Screenpipe CLI to query Enterprise team activity or safely preview, deploy, and schedule managed team Pipes. Injected only by the Enterprise app for active admins.

Authoring Fern MDX documentation pages for the Opik docs site, plus release-note and changelog routing. Use when writing or updating pages under apps/opik-documentation/documentation/fern/, drafting PR descriptions, or picking the right changelog surface.

Build or update an Opik integration that lives OUTSIDE this repo — a standalone opik-* package (e.g. opik-openclaw, opik-claude-code-plugin) or Opik support contributed into a third-party project (e.g. LiteLLM, Dify). Use ONLY when the user names an external repository or external package as the target; for integrations under sdks/ use the opik-integrations skill instead.

Use this skill to create fillable Word forms (.docx) with real Content Controls (SDT) + legacy FormField checkboxes + MERGEFIELD mail-merge placeholders + document protection. Trigger on: 'fillable form', 'form fields', 'content controls', 'SDT', 'word form', 'fill in', 'only editable fields', 'protect document', 'onboarding form', 'HR intake', 'survey template', 'contract / SOW template', 'mail-merge template', 'compliance checklist', 'medical intake questionnaire'. Output is a single .docx where specific fields are editable and the rest is locked. This skill is INDEPENDENT, not a scene layer on docx — payload is `w:sdt` + `w:ffData` + `w:fldChar` + `documentProtection`, none of which docx base skill covers. Do NOT trigger for regular reports, letters, memos, academic papers, pitch decks, or any document with no user-fillable fields — route those to officecli-docx or its scene layers.

Use this skill any time a .pptx file is involved -- as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file; editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions 'deck', 'slides', 'presentation', 'pitch', or references a .pptx filename.

This skill should be used when the user asks to "用 tmux 做真实测试", "保存 tmux 日志", "像真实用户一样测试 Qwen", "生成可复查的 TUI 测试报告", "测试 slash command 交互", or requests a tmux-based real user E2E run with complete readable logs. It guides real TUI usage with step-by-step capture-pane snapshots rather than ANSI raw pipe logs.

Review recent code changes for reuse, code quality, and efficiency, then directly apply straightforward cleanup improvements. Use when the user wants a post-implementation cleanup pass, pre-PR polish, or asks to simplify/refine recent changes. Invoke with `/simplify` or `/simplify <focus>`.

Coordinate a small team of Qwen Code teammates with enforced read-only workers, an optional worktree-pinned writer, shared tasks, peer messages, and existing Agent View tabs. Invoke explicitly with /coordinate.

Review a PR the way a seasoned maintainer would — build an independent model before seeing the diff, investigate the implementation and history, verify material claims, and return a concise briefing. Use for any PR review, self-review, or re-review.