菜单
SkillsMP 已收集 qihangzhang-272/agent-skill-library 中的 50 个 Skill。打开任一 Skill 可查看来源和详情。
已展示 40 / 50 个已收集 Skill。
MUST USE when user wants to 调研/research/搜索/search/查/找/look up anything on the internet — e.g. 全网调研 X / 帮我调研一下 X / 查一下 X / 搜搜 X / 看看大家怎么评价 X / X 上有什么讨论 / research this topic。 Also MUST USE when user mentions any platform or shares any URL/链接:…
Analyzes article structure, identifies positions requiring visual aids, generates illustrations with Type × Style × Palette three-dimension approach. Use when user asks to "illustrate article", "add images", "generate images for article", or "为文章配图".
原文语言：英语
Knowledge comic creator supporting multiple art styles and tones. Creates original educational comics with detailed panel layouts and batch-capable image generation. Use when user asks to create "知识漫画", "教育漫画", "biography comic", "tutorial comic", or…
原文语言：英语
Compresses images to WebP (default) or PNG with automatic tool selection. Use when user asks to "compress image", "optimize image", "convert to webp", or reduce image file size.
Generates article cover images with 5 dimensions (type, palette, rendering, text, mood) combining 11 color palettes and 7 rendering styles. Supports cinematic (2.35:1), widescreen (16:9), and square (1:1) aspects. Use when user asks to "generate cover image",…
原文语言：英语
Create clear, reproducible diagrams for architecture, flow, sequence, structure, state, and conceptual relationships. Use whenever the user asks for 架构图、流程图、关系图、状态图、decision tree、diagram or any visual representation of structure/logic/process. For…
原文语言：英语
Formats plain text or markdown files with frontmatter, titles, summaries, headings, bold, lists, and code blocks. Use when user asks to "format markdown", "beautify article", "add formatting", or improve article layout. Outputs to {filename}-formatted.md.
原文语言：英语
AI image generation with OpenAI GPT Image 2, Azure OpenAI, Google, OpenRouter, DashScope, Z.AI GLM-Image, MiniMax, Jimeng, Seedream, Replicate and Agnes APIs. Supports text-to-image, reference images, aspect ratios, and batch generation from saved prompt…
原文语言：英语
Generate professional infographics with 21 layout types and 22 visual styles. Analyzes content, recommends layout×style combinations, and generates publication-ready infographics. Use when user asks to create "infographic", "信息图", "visual summary", "可视化", or…
原文语言：英语
Converts Markdown to styled HTML with WeChat-compatible themes. Supports code highlighting, math, Mermaid (rendered to PNG via headless Chrome), PlantUML, footnotes, alerts, infographics, and optional bottom citations for external links. Use when user asks…
原文语言：英语
Posts content to WeChat Official Account (微信公众号) via API or Chrome CDP. Supports article posting (文章) with HTML, markdown, or plain text input, and image-text posting (贴图, formerly 图文) with multiple images. Markdown article workflows default to converting…
原文语言：英语
Apply the confirmed “航的杂谈地图” personal typography catalog to finalized Markdown, generate WeChat-compatible HTML, and create the read-only pre-publication preview used before draft publishing. Use when the user asks for 启航排版、个人公众号风格、个人样式目录、HTML 预览 or wants to…
Use when a workflow or skill needs to align with the library's operating principles: how to think, communicate, handle uncertainty, avoid empty output, and decide when to follow or challenge the user. This is a passive reference library, not a global rule…
Use when looking up curated external GitHub skill sources, deciding whether to install an external skill, or replacing vendored external skill folders with lightweight source references.
<何时触发 + 做什么。写得具体、略"推"一点以利触发。不要写"用于 XX 任务"这种废描述。>
Use when adding a skill, a new domain, bootstrapping the skill library from scratch, or extending principles. Generates spec-compliant plugin/skill skeletons, enforces the on-disk delivery protocol, and keeps the architecture self-contained and iterable.…
Route the library's proven workflows. Use when the user asks to run, package, choose, or maintain workflows for agent-reach research to public-account writing, Baoyu layout, visual-plan-driven production, HTML preview and WeChat draft publishing;…
原文语言：多语言混合
Use outside a stable Run when looking up curated external GitHub skill sources, deciding how to localize one, or replacing vendored external skill folders with lightweight source references.
原文语言：英语
Use when a candidate from the user's Agent Skill Library, another local repository, GitHub, an MCP, Prompt, Agent, API, script, or service must become a complete local Trial Skill before it can be called, tested, adopted, or inserted into a Personal Workspace…
原文语言：英语
Use when several adopted local Skills have formed a repeatable end-to-end sequence in real Cases, or when the user explicitly asks to preserve a proven way of working as a stable scenario Workflow with an outcome Benchmark.
原文语言：英语
Use when a Personal Workspace, capability lifecycle, source record, Host projection, inbox, or surrounding Case layout may be disorganized, duplicated, stale, or crossing its intended boundary, especially when the user asks for an inspection before any…
原文语言：英语
Use when a Personal Workspace may lack a capability, when choosing between an adopted local Skill, the user's Agent Skill Library, an existing candidate, or an internet source, or when asked to find a Skill, MCP, Prompt, Agent, API, or other reusable…
原文语言：英语
Use when the user explicitly evaluates an outcome, corrects a method, states a lasting preference, asks to upgrade a local capability, or when a registered upstream version should be considered as a candidate update to the Personal Workspace.
原文语言：英语
Use when a person is creating a new Personal Workspace, converting an existing folder into one, or asking how the same evolving local Skill environment will be shared by Codex App and Claude Code.
原文语言：英语
Use when the user gives a short outcome rather than naming Skills, asks for an end-to-end result in a complex scenario, or expects the current Codex App or Claude Code session to keep improving the work until the requested outcome is met.
原文语言：英语
从 BP / Pitch Deck 商业判断逻辑出发，对 AI 产品进行完整评估。 触发方式：- "分析一下 [产品名]" - "评估 [产品]" - "[产品名] 怎么样" - "帮我看看这个 AI 产品" - "这个产品值得关注吗" 输出：11 段 BP 逻辑链评估 → 综合判定（好案例/反面教材/待观察）+ 最强论点 + 最弱缺口 + 建议叙事线。 分析前必须判断是否需要读取 references/；命中商业模式、Data Agent、叙事审计等场景时先读取对应 reference，再输出判断。
Use before generating covers, article illustrations, infographics, architecture diagrams, or knowledge comics for Chinese public-account articles. Trigger whenever visual assets feel too diagrammatic, too AI-polished, emotionally empty, overloaded with…
Use when the user needs a final quality-control pass on an investment-banking pitch deck, client presentation, company profile, financial document, or capital-markets artifact for number consistency, narrative alignment, language, and visual readiness.
原文语言：英语
Use when the user needs a company tear sheet, company one-pager, fact sheet, company snapshot, corporate-development target profile, sales meeting brief, or investment-banking strip profile for a named public or private company.
原文语言：英语
Run the library's AI-native product judgment inside the investment workflow. Use after facts are collected and before market, unit economics, scorecard, DD, or IC memo writing. This skill evaluates whether the AI product is genuinely AI-native, commercially…
原文语言：多语言混合
Use when the user needs to populate, refresh, or quality-check an existing investment-banking PowerPoint pitch deck template using approved financial, transaction, market, or company handoffs.
原文语言：英语
Use when the user needs a sourced financial chart pack for public-equity initiating coverage, valuation, company performance, market structure, scenarios, or an institutional research report, and approved research, model, and valuation handoffs already exist.
原文语言：英语
Build the market and competitive landscape layer for an AI investment memo. Use when an IC memo, investment memo, product investment report, or AI case review needs market sizing, competitor mapping, positioning, moat assessment, bull/base/bear scenarios, or…
原文语言：多语言混合
Generate diligence priorities, red flags, data-room requests, management questions, expert-call questions, and one-vote-veto checks for an AI investment memo. Use after product, market, unit economics, scorecard, and valuation nodes have produced preliminary…
原文语言：多语言混合
Use when the user explicitly needs an auditable XLSX financial model, three-statement forecast, DCF or LBO workbook, model sensitivity, or spreadsheet audit before valuation. Do not use for narrative-only valuation or when verified historical inputs and…
原文语言：英语
Write the final library-standard investment committee memo from completed investment workflow node outputs. Use only after fact collection, AI product judgment, competitive landscape, unit economics, scorecard, valuation, DD, and thesis tracking have produced…
原文语言：多语言混合
Collect and organize source-backed facts before AI investment analysis. Use before IC memo, investment memo, AI case investment review, DD prep, or visual investment report when product, company, financing, customer, pricing, competitor, GitHub, or market…
原文语言：多语言混合
Convert AI product judgment and fact packs into the library's investment scorecard layer. Use for AI / OSS / infrastructure investability scoring, macro gate, deal screening, one-vote veto checks, pass / watch / recommend decisions, and scorecard handoff into…
原文语言：多语言混合
Turn an AI investment memo into a falsifiable thesis, catalyst calendar, watch triggers, KPI monitoring plan, and update process. Use when the user wants to track an AI company, open-source project, investment case, portfolio company, or watchlist item after…
原文语言：多语言混合
Analyze business model quality, revenue quality, customer economics, AI readiness, and value creation for AI investment cases. Use when an IC memo needs ARR, NDR, LTV/CAC, CAC payback, gross margin, inference cost, pricing, revenue concentration, pilot…
原文语言：多语言混合