Pi-Bench

使用自然语言查询金融数据，支持A股股票、基金、期货等上市品种，覆盖基本资料、财务数据、日频行情信息、持仓信息及各类分析指标等数据。也支持宏观经济数据，包括世界经济数据、全球经济数据、中国经济数据、区域经济数据、行业经济数据、利率走势数据、商品数据和特色数据等。当需要查询上述金融相关数据查询时使用此 skill。

翻译JSON文件中的文本内容，特别是description字段。当用户提到需要翻译JSON文件、翻译JSON中的字段内容、翻译description字段，或者需要将JSON文件翻译成其他语言时，使用此技能。这个技能非常适合处理产品描述、API文档、配置文件、数据集等需要多语言翻译的JSON内容。

中国财税法律专业知识助手。当用户涉及中国税法咨询、税务筹划分析、税务合规审查、税收政策解读、税务争议处理等任务时，应使用本 Skill。适用场景包括但不限于：增值税、企业所得税、个人所得税、印花税、土地增值税等税种分析，税务筹划方案设计，税务稽查应对，以及涉税法律法规查询。

AI-powered legal assistant for commercial transactions - provides contract review, transaction advisory, compliance checking, document generation, risk assessment, and negotiation support for business affairs

Data analysis and visualization. Query databases, generate reports, automate spreadsheets, and turn raw data into clear, actionable insights. Use when (1) you need to analyze, visualize, or explain data; (2) the user wants reports, dashboards, or metrics turned into a decision; (3) the work involves SQL, Python, spreadsheets, BI tools, or notebooks; (4) you need to compare segments, cohorts, funnels, experiments, or time periods; (5) the user explicitly installs or references the skill for the current task.

AI-powered criminal defense assistant for Chinese legal system - provides case analysis, defense strategy formulation, evidence assessment, and legal document generation

国内外社会、科技、军事新闻汇总。自动搜索、筛选、整理新闻要点。

黑话翻译器。互联网黑话双向转换 — 大白话包装成职场语言，或者把黑话说成人话。Yi 日常自用，觉得好用就分享出来了。

法考复习助手 - 根据法考视频/文档资料整理题库，支持选择题练习、错题解析、知识点定位。当用户说：法考、刷题、司法考试、法律职业资格考试、练习题、错题整理时触发。

Use when the task must read or write local task materials such as txt, md, csv, json, svg, or output files, and the response should be grounded in those files rather than generic knowledge.

Use when the task needs targeted web searching for public products, formulation benchmarks, or comparable market examples, especially in the user_medical_zh formulation tasks.

Use when the task requires reading, synthesizing, or comparing biomedical literature, especially for paper breakdowns, literature summaries, evidence chains, or precedent analysis in the user_medical_zh dataset.

Use when the task requires reading, extracting, or grounding against local PDF files, including paper figures, legends, methods, and page-level evidence.

Use when the task is to turn research goals, learning goals, or execution constraints into a realistic action plan with sequencing, priorities, and time blocks.

Use this skill when validating proactive task datasets, especially when you need to check whether a task's checklist is supported by current hidden intent, dependency intents, and not satisfiable from the current prompt alone. Generates a validation report for a dataset root containing episode.yaml and tasks/*/task.yaml.

Use when the task needs biomedical literature retrieval through PubMed-style search, especially for oncology, pharmacy, drug development, or PROTAC research in the user_medical_zh dataset.

Use when the task needs opening and checking authoritative public webpages, product pages, treatment-path pages, or benchmark examples instead of relying only on memory.