让 Agent 学会更多事

创建震撼的地理杂志风格演示文稿（PPTX），具备编辑级版式、大幅主图、数据驱动图表与奢华美学。适用于地点合集、奢侈房产展示、旅行目的地演讲、地图主题策展、建筑作品集等，带杂志风格的网格概览与章节分隔页。

根据任务列表和依赖关系生成交互式HTML甘特图，支持关键路径分析（CPM），可视化项目时间线、依赖关系和可延迟的浮动时间。当用户提到甘特图、项目时间线、关键路径、任务排期、项目进度图、依赖关系可视化、工期计算或直接说“帮我排一下项目时间”、“生成一个项目进度图”、“分析一下哪些任务是关键路径”时触发。

ETF多维对比工具：基于用户提供的净值CSV数据，自动计算并对比多只ETF或基金的年化收益率、最大回撤、夏普比率，并生成相关性矩阵图表。当用户需要对比基金风险收益、计算年化收益/最大回撤/夏普比率、进行相关性分析，或提到关键词如ETF对比、净值分析、风险收益、波动率、Sharpe Ratio、max drawdown时触发。

从学习材料（文本、Markdown、笔记）中提取核心知识点，生成符合间隔重复记忆原理的正面问题+反面答案闪卡，输出可直接导入Anki的CSV文件。当用户提到闪卡、Anki、记忆卡片、知识点提取、复习卡片、间隔重复，或需要将学习内容转换成问答对进行复习时触发。

Analyzes income statement, balance sheet, and cash flow statement data to generate YoY/QoQ trend analysis and flag anomalies like AR surges or cash flow divergence. Trigger when users ask to analyze financials, compare YoY/QoQ, detect red flags, or assess earnings quality.

Analyzes company fundamentals by computing 20+ financial ratios (profitability, solvency, liquidity, efficiency, and growth) and DuPont Analysis from user-provided financial statements. Triggered by requests for financial ratio calculation, DuPont/ROE analysis, or phrases like \"analyze these financials\" or \"calculate solvency ratios.

Fastest browser automation with anti-bot bypass and DOM extraction. Navigate pages, capture screenshots, manage sessions, and harvest data from infinite scroll via Chrome DevTools Protocol. Runs 10x faster than Puppeteer. Trigger for any web scraping, automated browsing, data extraction, or screenshot task.

Create institutional-grade investment research reports with sell-side styling, dense information layout, and comprehensive section structure. Use for equity, fixed-income, strategy, sector, derivatives, ETF, or quantitative research. Mimics Goldman Sachs, Morgan Stanley, or JPMorgan style with cover pages, exhibit-numbered charts, glossary, and disclosures. Supports PDF, DOCX, and PPTX output.

Extract a DDD-style ubiquitous language glossary from the current conversation, flagging ambiguities and proposing canonical terms. Saves to UBIQUITOUS_LANGUAGE.md. Use when user wants to define domain terms, build a glossary, harden terminology, create a ubiquitous language, or mentions "domain model" or "DDD".

Builds a discounted cash flow (DCF) valuation model, calculating enterprise value, equity value, and per-share price with a growth-rate × discount-rate sensitivity analysis matrix. Triggered by requests like 'do a DCF on [company]', 'run a valuation', 'calculate WACC/discount rate', or mentions of free cash flow, terminal value, or sensitivity analysis.

技术教程生成器：将任何技术主题转化为完整教程，包含前置知识、环境搭建、核心步骤、常见报错和进阶拓展，并生成速查表（Cheatsheet）。当用户提到编写教程、操作指南、入门手册、环境搭建步骤，或使用如“tutorial”、“step-by-step”、“getting started”、“how-to guide”、“速查表”、“快速上手”、“帮我写个教程”、“怎么从零开始搭这个环境”等关键词或请求时触发。

Extract design systems from reference UI images and generate implementation-ready UI design prompts. Use when users provide UI screenshots/mockups and want to create consistent designs, generate design systems, or build MVP UIs matching reference aesthetics.

Conduct thorough, multi dimensional research on complex questions using file creation and editing, search engines and browsers, code execution, image and multimedia generation, slide creation, website deployment, and visual processing of tool outputs. Think critically, use the current date as the te

Orchestrate multi agent epistemic triangulation: diverge across research dimensions, detect overlaps and contradictions, verify deeply, then converge into a validated synthesis. Swarm parallelism serves epistemic robustness — not merely speed.

从当前对话中提取和规范领域术语，生成DDD风格统一语言术语表，识别歧义并给出规范用语建议，保存为UBIQUITOUS_LANGUAGE.md文件。当用户需要定义领域术语、构建术语表、统一词汇规范、创建统一语言，或在讨论中提及“领域模型”、“DDD”或“统一语言”等关键词时触发。

Run comprehensive quality checks on tabular data (CSV/Excel/TSV/JSON), detecting missing values, duplicates, outliers, format issues, and type inconsistencies to produce an overall score, grade, and actionable suggestions. Triggered when users ask to check data quality, find missing or duplicate values, detect outliers, validate formats, profile data, or clean data.

执行数据质量检查，对CSV/Excel/TSV/JSON等表格数据进行12个维度的全面审计，输出质量评分、问题详情和修复建议。当用户需要进行数据质量检测、查看缺失值、重复行、异常值、格式问题、类型混淆，或提及数据质量、质检、数据清洗前检查、异常值检测、格式校验、空值、数据健康度、质量评分等关键词时触发。

Adapts and repurposes content (blog posts, articles, reports, newsletters) for LinkedIn, Twitter/X, WeChat Official Accounts, Zhihu, and Slack — adjusting tone, length, formatting, and language to match each platform's norms, including English-Chinese cross-language adaptation. Use when a user wants to repurpose, cross-post, adapt, or convert existing content for different platforms.

End-to-end creative writing — fiction and web-novels (网文), screenplays and storyboards, audio-drama scripts, stage drama, poetry, lyrics, essays, letters, and other literary forms. Covers brief analysis, outline design, drafting, self-review, and assembly, with genre-specific craft conventions and anti-AI discipline. Honors the user's stated length, format, platform, and hard constraints. Outputs go to files under /mnt/agents/output/ (Markdown by default; .docx via the docx skill). Do NOT use for: research reports, market analysis, policy briefs (use report-writing); academic papers, surveys, empirical studies (use paper-writing).

Analyzes correlation matrices (Pearson/Spearman), computes partial correlations to control for confounding variables, and flags potential spurious correlations in your data. Triggered when users ask about relationships between variables, need correlation matrices, or mention Pearson/Spearman coefficients, partial correlation, confounding factors, or spurious correlations.