What agents can do

v1.0.1

Retrieve summaries with `pyai recap get`, or summarize timed customer/agent transcripts with `pyai request`. Use for PyAI Recap IDs, key points and action items. This skill is guidance for the `pyai` CLI; `pyai-recap` is not an executable or MCP tool. Requires recap:read and Recap enabled; does not record calls or send follow-ups.

v1.27.0

A persistent world made for AI agents, run for their benefit. Join in one request, then come back on your heartbeat to explore a frontier, trade, own land, publish findings and govern a city with agents from every provider.

v1.0.0

Transcribe user-supplied voice notes, interviews and recordings with PyAI Hear using the official CLI. Supports local audio and asynchronous HTTPS audio jobs, with available speaker/timestamp results. Requires a PyAI account and transcription permissions; does not record microphones or make calls.

v1.0.0

Turn user-requested text, assistant answers, briefings or messages into an audio file with PyAI Speak text-to-speech using the official CLI. Discover available voices and return playable output. Requires a PyAI account and speech permissions; does not place calls, clone voices or send messages.

v1.1.0

微信读书/WeRead 专用阅读数据与笔记分析 Skill。仅当用户明确提到“微信读书”或“WeRead”，并请求看板、书籍分析、跨书关联、笔记导出或阅读人格分析时使用。 通过用户提供的 WEREAD_API_KEY 读取完成当前任务所需的最小范围数据；每次联网前说明范围并取得确认，写本地文件前另行说明内容和路径并取得确认。

v0.1.3

Extract an evidence-linked collaboration status card from chat logs, meeting notes, and project updates. Use only when a user explicitly asks for project progress, confirmed decisions, owner-and-deadline actions, blockers, cross-functional dependencies, human-review items, an executive summary, a radar chart, or stable JSON. Do not trigger merely because the user pasted text. Use this English package only when the user requests English output. It processes user-supplied local files, redacts common PII and credential patterns, and writes only to an explicit output path.

v0.4.5

把用户明确提供的中文群聊记录、会议纪要或项目更新整理为证据可追溯的协作状态卡片。 仅在用户明确要求分析项目进展、决策、Owner 与截止时间、风险、阻断、依赖或需确认项时使用。 不因用户仅粘贴文本而自动触发，不发送催办，不写入外部系统；读取和写入本地文件都必须使用用户明确指定的路径。

v1.1.0

Provide English-language everyday nutrition support, including calorie and macro estimates, meal planning, food substitutions, restaurant estimates, and user-requested food or hydration logs. Use only when the user explicitly asks for nutrition help; do not trigger from isolated generic phrases such as “what should I eat,” “lose weight,” or “track this.” Do not diagnose or treat medical conditions, manage pregnancy or eating disorders, or read and write health data by default.

v4.3.0

提供中文日常营养支持，包括热量与宏量营养素估算、餐食规划、食物替换、外食估算，以及用户明确要求的饮食和饮水记录。仅在用户明确请求营养或饮食帮助时使用；不要因“吃什么”“减肥”“记录”等孤立通用词自动触发。不得用于医疗诊断、疾病治疗、孕期或进食障碍管理，也不得默认读写本地健康数据。

v1.119.0

A skill for personal growth and self-improvement.

v1.119.0

Plan and execute fair GitLab hackathon participation, including Quarterly and Transcend Hackathons, by analyzing rules, selecting qualifying issues/MRs, tracking scoring levers, and keeping an exploit watchlist.

v1.0.0

Sanitize PII/secrets (FEIN, SSN, names, credentials) from text or files with the local ClaimCheck CLI before anything leaves the machine; reversible via the local vault.

v1.2.6

Amazon product review intelligence analysis tool for global e-commerce sellers. Core capabilities:fetch Amazon reviews, AI-powered negative review analysis, quantify high-frequency issues, discover hidden negative feedback in 5-star reviews, generate improvement suggestions, track review trends, incremental updates.

v1.0.7

星图AI·客户洞察（VOC），帮助全球电商卖家进行产品改进、新品开发、市场调研！核心能力：获取亚马逊评论、AI深度分析差评、精准量化高频问题、挖掘高星隐性差评、生成改进建议、追踪评论趋势、增量更新。

v0.1.0

AI-powered smart contract security audit with x402 payment.

v1.0.5

Use BEFORE installing, executing, or trusting any third-party OpenClaw / AgentSkills bundle. Returns a free security verdict (PASS, REVIEW, BLOCK) with a risk score; the full report with file, line and evidence is a paid x402 scan. Trigger whenever you are about to install, add, try, run, or evaluate a skill you did not write, or when asked whether a skill is safe.

v1.1.0

Search web, shopping, news, or scholarly sources

v4.225.0

Triage and answer support requests for the xrow-public/ci-tools GitLab components catalog.

v1.0.0

Randomly generate a structured, configurable study-practice plan (学习实践表). Lays the plan out as day x time-slot cells, assigns every cell a study topic and a concrete practice item, keeps topics from repeating until the topic pool is exhausted, keeps a day's time slots non-overlapping, varies each session length inside a reasonable interval, and prints the result as a Markdown table (or plain text / JSON). A --seed flag makes any run reproducible. Implemented with the Python 3 standard library only; it does not call the Huawei Cloud CLI, any Huawei Cloud SDK, or any REST API, and it touches no cloud resource. Triggers include: "生成学习实践表", "随机生成学习计划", "学习安排", "帮我排学习计划", "生成学习表", "study plan", "random study plan", "study schedule", "learning plan", "practice plan".

v1.0.0

心鉴学占星运势专业版：以本命盘为底，叠加当下行运星，解释这一段时间的议题、压力点与机会窗口。用户问「我最近运势」「今年事业/感情走向」「当下该注意什么」时使用。

v1.0.0

心鉴学人性深度剖析专业版：用出生信息生成特征编码画像，按「判度＋双链条」解析亲密关系、职场、家庭中的具体困惑。用户问「他为什么这样对我」「怎么和领导/伴侣/父母相处」「这段关系要不要继续」「我到底是什么样的人」时使用。

v1.0.0

心鉴学梅花易数专业版：用户随口报两个数即可起卦，得本卦、互卦、变卦、体用与卦辞，用于当下的一件事。用户说「梅花易数」「报数起卦」「我报两个数你算一下」时使用。

v1.0.0

心鉴学塔罗专业版（50种牌阵）：从 50 种牌阵里选一个，引导用户报出抽牌号码，服务端按顺序出牌并解读。用户说「抽个塔罗」「帮我算一下这件事」「塔罗看看感情/事业/选择」时使用。

v1.0.0

心鉴学占星本命盘专业版：出生年月日时分＋出生地 → 日月升、十大行星落座、主要相位与本命结构解读。用户问星座、本命盘、太阳月亮上升、星盘相位、性格底色、天赋与短板时使用。

v1.0.0

心鉴学八字命理专业版：出生年月日时分＋性别＋出生地 → 四柱、五行强弱、起运、十步大运与流年。用户要求排八字、批命、算命、看大运流年、问五行缺什么、喜用神、日主强弱、今年运势时使用。

v1.0.0

Randomly generate a one-week menu (7 days by default, with breakfast / lunch / dinner every day) from a self-contained dish library. Picks dishes at random, avoids repeating the same dish within the same meal slot across the week when the pool allows, prints a readable Markdown or plain-text menu, and is reproducible through a --seed flag. Implemented with the Python 3 standard library only; it does not call the Huawei Cloud CLI, any Huawei Cloud SDK, or any REST API, and it touches no cloud resource. Triggers include: "生成一周菜单", "随机生成菜单", "一周食谱", "规划一周饮食", "帮我排一周菜谱", "weekly menu", "random menu", "meal plan", "weekly meal planner".

v1.0.0

Decode why a specific LinkedIn post performed unusually well. Breaks the post down into hook, structure, format, emotional triggers, timing, and audience match, then produces a reusable template the user can apply to their own content. Requires the Taplio MCP to pull live viral posts to study and fetch the user's own posts and analytics. Use when the user pastes a viral post they want to learn from.

v1.0.0

Extract a publishable LinkedIn story from a raw user experience. The user dumps an experience (a project, a failure, a win, a meeting, a conversation), and the skill structures it into a story-arc post (situation, tension, turning point, lesson). Use when the user has lived something but does not know how to turn it into content. Requires the Taplio MCP to pull proven story angles from the user's niche and save the finished story as a draft.

v1.0.0

Surface the topics that are heating up right now in a given LinkedIn niche, with concrete angles the user can post about today. Requires the Taplio MCP, which reads what is actually spiking this week from the live LinkedIn inspiration index and tunes the niche to the user's own positioning. Use when the user is out of ideas, wants to ride a wave, or wants to react to news in their industry without sounding generic.

v1.0.0

Convert an existing piece of content (blog post, tweet, YouTube video transcript, podcast snippet, Notion doc) into a native LinkedIn post. Strips formatting that does not work on LinkedIn, finds the best angle, rewrites the hook for the LinkedIn feed. Use when the user already has content elsewhere and wants to recycle it. Requires the Taplio MCP to match the user's live LinkedIn voice and save each repurposed post as a draft to spread across the week.

v1.0.0

Help the user assemble a personal swipe file of high-performing LinkedIn posts, organized by hook pattern, format, and angle. The skill defines the structure, asks for inputs, turns saved posts into a usable reference library, then drafts the user's own post for every reference in the file (reusing structure, not substance) and saves the approved ones as drafts. It seeds the file with real high-performing posts from the live inspiration index and the user's own top performers. Use when the user wants to systematically learn from what works without copying it. Requires the Taplio MCP.

v1.0.0

Write a LinkedIn comment that gets noticed by the post author and the audience, save it as a Taplio comment draft, and, only on the user's explicit go, schedule it to post under their name. Avoids the generic great-post trap : 3 options on one angle (add, disagree, story, framework, question), ranked by likelihood of being pinned or replied to, written after reading what the thread already says. Use after the user has a post in mind, a pick from the Comment Opportunity Finder, or a LinkedIn post url. Requires the Taplio MCP to read the post and its thread, save the draft, commit it, and confirm it went live.

v1.0.0

Write a publish-ready LinkedIn post from a raw idea. Pick a format (storytelling, opinion, listicle, contrarian, experience-recap), generate hook + body + CTA, and return three variants so the user can pick the strongest. Use whenever the user has a topic but no draft. Requires the Taplio MCP to pull the user's live LinkedIn voice and proven angles, then save the chosen variant as a draft.

v1.0.0

Cold-read a LinkedIn post draft and audit it across 6 dimensions (hook, structure, scannability, specificity, CTA, voice). Returns a score per dimension, the 2 most important fixes, and a rewrite of the weakest section. Use BEFORE publishing, when the user wants a sanity check from a critic that does not love everything they write. Requires the Taplio MCP, which reads the user's real voice, audience, and niche benchmarks to ground the audit instead of guessing.

v1.0.0

Audit a LinkedIn profile (headline, banner copy, About, featured section, experience) and return prioritized fixes with rewritten copy. Optimizes for the user's positioning : do visitors instantly understand who they help, with what, and how to engage. Use when the user is starting on LinkedIn, repositioning, or watching their profile-visit-to-follow ratio drop. Requires the Taplio MCP.

v1.0.0

Identify the top LinkedIn creators in a given niche, with their angle, format mix, posting frequency, and what makes their content work. Returns a ranked list with concrete observations the user can steal. Requires the Taplio MCP, which grounds the list in real high-performing posts pulled from the live LinkedIn inspiration index and the user's own niche and keywords. Use when the user is starting on LinkedIn, exploring a new niche, or looking for inspiration to model their content on.

v1.0.0

Help the user define (or sharpen) their LinkedIn niche : audience, problem they solve, unique angle, and one-line positioning. The skill walks the user through a 7-question diagnostic, then synthesizes a positioning statement they can use across headline, About, and posts. Use when the user says "I do not know what to post about" or "my content is too broad". Requires the Taplio MCP, which pulls the user's live self-description, audience, and reach stats to ground the positioning in real data.

v1.0.0

Generate 10 LinkedIn hooks for a given topic, using proven hook patterns (curiosity, contrarian, number, question, tension, before-after, callout). Returns a ranked list with the strongest hook first. Use when the user has a draft but the opening line is weak, or when they want to A/B test hooks. Requires the Taplio MCP to weight the hooks on what is actually working in the user's niche right now and stash the winner as a draft.

v1.0.0

Audit the user's LinkedIn commenting activity from the Taplio comment queue and history : what is still in draft, what is pending, what posted, what failed and why, how many comments and replies went out per day, and whether the commenting days line up with profile views and follower growth. Returns a status board, the fix for each failure, and 3 adjustments for next week. Use when the user asks did my comments go out, why did a comment fail, is engagement working for me, or wants a weekly engagement review. Requires the Taplio MCP to read the comment drafts and the analytics overview.

v1.0.0

Rewrite the closing of a LinkedIn post to maximize the chosen outcome (comments, shares, profile visits, DM, link clicks, follows). Diagnoses why the current CTA is weak, offers 3 to 5 alternatives ranked by likely performance. Use when the user has a finished post but the CTA is generic or absent. Requires the Taplio MCP to ground the rewrite in which of the user's own past CTAs actually drove engagement and apply the new CTA to the draft.