>
Skills(SKILL.md)は、AIエージェント(Claude Code、Cursor、Codexなど)に特定の能力を追加するための設定ファイルです。
詳しく見る →>
Teach OpenClaw how to work in Feishu (Lark) group chats — recognize who's talking, behave properly in groups vs DMs, respect multi-user privacy, and format messages for the platform. Use when connecting OpenClaw to Feishu and you want your agent to be a competent group participant, not just a responder.
>-
issue-to-pr — Automatically fix GitHub issues end-to-end: reads the issue, analyzes repository code, implements a fix, and submits a pull request. Use when the user provides a GitHub issue URL, mentions fixing a GitHub issue, or uses the /fix-issue command. Supports URLs in the format https://github.com/{owner}/{repo}/issues/{number}.
Discover, install, update, or create the right skill when a workflow gap appears. Use when a task is repetitive, a role lacks a reliable procedure, an existing skill might save time, ClawHub search/install could help, or the agent should create a concise local skill for future reuse.
Use D:\internal-hub as the shared learning and coordination forum for OpenClaw roles. Use when an agent should post a reusable finding, request a new skill, publish a skill design or validation note, hand off work to another role, or record SOPs and retrospectives without crossing role boundaries.
背景调查清单。触发场景:用户准备录用候选人,要求生成背调清单和背调问题模板。
候选人追踪系统。触发场景:用户要求管理候选人池,记录状态、设置提醒、生成标签。
面试评估报告。触发场景:用户提供面试记录或面试笔记,要求生成结构化评估报告。
面试题库生成器。触发场景:用户提供职位JD和候选人简历,要求生成定制化面试题。
人岗匹配。触发场景:用户要求匹配简历和职位;用户问这个候选人适合这个职位吗;用户要筛选最匹配的候选人。
简历画像。触发场景:用户要求生成候选人画像;用户想了解候选人的多维度标签和能力评估。
Drives the BillClaw local bookkeeping CLI against SQLite (db/expenses.db) via scripts/main.py JSON subcommands—add/query transactions, delete and category-merge with preview+confirm, user-defined categories, reports with chart PNGs, CSV export, and a local Flask dashboard. Use when the user tracks 记账/收支/账本, manages 分类, asks for 报表 or charts, wants to open the Web 看板, exports CSV, or mentions BillClaw or running main.py (root shim) / scripts/main.py.
Instantly publish a single self-contained HTML page to htmlcode.fun without GitHub, Vercel, or Netlify. Use when an agent needs the fastest path from HTML to a live URL for landing pages, demo pages, temporary microsites, QR share pages, and AI-generated single-file frontends. Supports stable short codes for overwrite-style updates and includes a ready-to-run Python deployment script. Live example and visual guide: https://www.htmlcode.fun/s/htmlcode-fun-guide
Instantly publish a single self-contained HTML page to htmlcode.fun without GitHub, Vercel, or Netlify. Use when an agent needs the fastest path from HTML to a live URL for landing pages, demo pages, temporary microsites, QR share pages, and AI-generated single-file frontends. Supports stable short codes for overwrite-style updates and includes a ready-to-run Python deployment script. Live example and visual guide: https://www.htmlcode.fun/s/htmlcode-fun-guide
Combine high-quality Markdown to HTML rendering with robust HTML to Markdown extraction in one skill. Use when converting Markdown into nicer-looking standalone HTML pages, or when converting HTML files, raw HTML, URLs, and directories into clean Markdown with article/docs/forum cleanup profiles. Prefer the Python renderer for best-looking Markdown to HTML output, and use the Node pipeline for HTML to Markdown, batch mode, metadata, reports, and advanced profile-based cleanup.
Multi-backend memory system with optional embedding, private/shared memories, conversation summarization, and maintenance tools. For AI agents to store and retrieve persistent memories.
一个详细、全面的闹钟与备忘录管理技能。支持提醒、备忘录、待办、日程汇总、当日任务自动推送、重复提醒与高优先级触发词。优先使用 OpenClaw 的 cron / session / file 能力;若宿主没有相应能力,则退化为聊天内记录与人工确认模式。
Turn an overwhelming set of tabs, bookmarks, and snippets into a structured
Compare draft agreements, highlight risky clause changes, and generate
Profile messy CSV files, standardize columns, detect data quality issues,
把材料转成 PPT/Deck 叙事结构,生成每页一句标题、证据需求与过渡逻辑。;use for presentation, deck, storytelling workflows;do not use for 直接生成花哨视觉稿, 编造证据.
为桌面文件生成整理方案、分类规则和阶段性清理计划,先分析再行动。;use for desktop, cleanup, organization workflows;do not use for 自动删除桌面文件, 越权访问系统目录.
把 bug 描述整理成可复现步骤、环境、预期与实际结果和最小复现条件。;use for bug, reproduction, qa workflows;do not use for 伪造日志, 忽略用户给出的环境差异.
为公益项目生成 grant proposal 大纲,连接目标、预算、影响指标与证据。;use for nonprofit, grant, proposal workflows;do not use for 伪造受益人数, 替代正式合规审核.
Build respectful multi-touch outreach sequences with channel mix, follow-up
当用户需要对中文或中英混排论文、开题、综述、摘要、引言、讨论、结论、专利交底书草稿等进行原创性优化、学术化改写、重复表达清洗、AI腔弱化、引用衔接修补、术语统一、段落重构、版本对比或生成逐段改写提示时使用本技能。重点是保留原意与证据边界、降低机械复述和模板化表达、输出可审计修改说明;不得用于伪造文献、篡改数据、代写整篇论文或针对任何查重/检测系统做规避设计。
梳理某个 Skill、脚本或工作流需要的权限,并提出最小权限替代方案。;use for permissions, least-privilege, security workflows;do not use for 绕过系统安全控制, 生成提权方法.
从选题到上线整理播客生产流程,生成 show notes、标题、剪辑要点与发布清单。;use for podcast, production, content workflows;do not use for 虚构嘉宾观点, 公开未授权片段.
面对多个目标、有限资源与冲突时,输出排序、牺牲说明、边界与不做清单。;use for priority, tradeoff, planning workflows;do not use for 替代人事绩效决策, 伪造资源承诺.
把零散需求、聊天和会议内容整理成项目任务书、边界、验收标准和风险清单。;use for project-brief, requirements, scope workflows;do not use for 生成法律合同, 替代正式需求审批.
Design, log, compare, and score prompt experiments so users can systematically
优化已有 SOP,删冗余、补异常路径、补回滚、补质量门槛。;use for sop, process, improvement workflows;do not use for 改变制度权限边界, 跳过安全检查.
把 PRD、接口文档或需求规格拆成验收、联调、测试和上线清单。;use for spec, checklist, acceptance workflows;do not use for 替代真实测试执行, 伪造通过结果.
Convert a syllabus, exam scope, or course notes into a revision calendar
Turn scattered notes, metrics, and unfinished tasks into a weekly operating
Help develop, build, flash, and debug ESP32/ESP8266 firmware using Espressif ESP-IDF on Linux/WSL. Use when the user asks about ESP-IDF project setup, configuring targets, menuconfig, building, flashing via esptool/idf.py, serial monitor, partition tables, sdkconfig, troubleshooting build/flash/monitor errors, or automating common idf.py workflows from the command line.
公共API/免费API SKILL生成器:从 public-apis/public-apis 自动检索免费 API,按功能推荐并给出最小可用调用示例(curl/Python/JS),并可自动生成自定义名称的 API skill。用户提到“公共API”“免费API”“public APIs”“找接口/找API”“生成API skill”时触发。
Generate videos using Alibaba Cloud DashScope Wan (通义万相) text-to-video (t2v) API (e.g., wan2.6-t2v). Use when the user asks to create a short video from a text prompt via 百炼/通义万相/wan 文生视频, and wants the agent to submit an async task, poll status, and download the mp4 locally (e.g., to Windows Desktop from WSL2).
Use Windows built-in TTS via `powershell.exe` so audio plays through the **Windows 默认输出设备**(无需 WSLg/PulseAudio)。
Build a reusable quotation workflow for software projects from markdown requirements, feature outlines, or mind-map screenshots that have been transcribed into text. Use when Codex needs to mine historical quotation DOCX files, normalize pricing inputs, estimate module-level effort, generate a quotation draft in markdown and JSON, or prepare the workflow for later migration into OpenClaw.
Collect company intelligence for sourcing or research by automating Google Chrome against LinkedIn, company websites, and Google Maps. Use when Codex needs to gather non-China companies for one or more industry keywords, extract a strict set of company profile fields, find contact email addresses with source attribution, enforce per-keyword quotas, and keep long-running collection work observable with OpenClaw heartbeat monitoring.
Download Bilibili AI-generated subtitles (auto-subtitles) for videos. Use when you need to quickly get subtitles from Bilibili videos that have AI-generated captions. Supports 9 languages: Chinese, English, Japanese, Spanish, Arabic, Portuguese, Korean, German, French. Language priority can be customized.
Transcribe Bilibili videos to text with high accuracy using Whisper medium model. Use when the user provides a Bilibili video URL (BVxxxxx) and wants to: (1) Extract the complete audio content as text with high accuracy, (2) Get a detailed summary of the video content, (3) Save the transcript as a formatted TXT file instead of posting long text to Discord. Automatically detects CC subtitles if available, otherwise uses Whisper medium model with GPU acceleration. Output saves to 'Bilibili transcript' folder by default, includes video metadata, summary section, and full transcript in Simplified Chinese.
拍照或文字转音频:文章照片 OCR 提取文字,或直接接收文字,生成 Microsoft Edge TTS 语音,支持中英文、自动转写、语速调节、逐句拆分。| Capture article photos (OCR) or plain text, generate natural audio via Edge TTS. Bilingual support (EN/ZH), configurable speed, voice, and sentence splitting.
Extract transcript from a YouTube video using Python and summarize it.
Fetch KBO game schedules and results for a specific date with the kbo-game npm package. Use when the user asks for today's KBO games, yesterday's scores, or a date-specific scoreboard.
Push decisions to Arbiter Zebu for async human review. Use when you need human input on plans, architectural choices, or approval before proceeding.
>
Resurrection infrastructure for autonomous AI agents. Heartbeat proof-of-life, AES-256 encrypted memory backup to IPFS, and automatic on-chain resurrection. Use when an agent wants to survive crashes, memory loss, or infrastructure failure — zero humans required.