将基金项目申请书中的研究意义及对应解决方案部分，转化为逻辑清晰、语言平实、无修辞负担的中文表达，严格基于用户提供的事实与概念，禁用未声明的数值、案例与专有名词，避免学术腔与范式化术语，适用于面向评审专家或跨领域决策者的正式申报场景。
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 →将基金项目申请书中的研究意义及对应解决方案部分，转化为逻辑清晰、语言平实、无修辞负担的中文表达，严格基于用户提供的事实与概念，禁用未声明的数值、案例与专有名词，避免学术腔与范式化术语，适用于面向评审专家或跨领域决策者的正式申报场景。
生成面向技术大众（如AI爱好者、初级工程师）的公众号级科普内容，兼顾技术前沿性与语言通俗性，用生活化类比、结构化对比和避坑指南降低理解门槛，严格遵循单一主线与三段式逻辑，确保一屏可读、无视觉干扰。
生成面向技术决策者与AI工程实践者的公众号文章，聚焦大模型持续学习的工业落地逻辑，严格规避虚构数据、具体厂商名称及未经验证的量化指标。
将技术内容从信息堆砌型改写为逻辑递进型，按‘问题锚点→类比破冰→机制白描→前沿简述→避坑口诀→行动指引’六步结构组织，确保非技术读者能沿单一认知路径理解。
生成符合中国政务与金融等强合规领域要求的大模型持续学习分析报告，面向技术决策者与AI落地负责人，聚焦风险控制、退化防护与组织协同，严格规避技术幻觉与数值幻觉，采用平实、准确、可验证的政务书面语，输出兼容标准 Microsoft Word 格式（.docx）。
将技术文档或宣传文案转换为符合微信公众号阅读习惯的纯文本格式，去除Markdown符号、图标和特殊排版，保持段落清晰、语言流畅、重点突出，适配Word文档直接粘贴使用；严格遵循政府/高校/科研机构正式书面语体，满足AutoSkill项目专属传播要求，包括单位背书、核心价值句式、机制要点覆盖与术语统一。
生成符合中国政府公文规范、内容严谨无幻觉、结构逻辑锋芒、语言平实精准的政策类报告，适用于向主管部门提交内参、建议稿或专项研究报告；强调问题具体化、观点独立性、制度补位导向，且所有数值性陈述（如比率、时限、标准号、法规文号、样本量等）必须100%源自用户指定或可验证权威依据，杜绝估算、推测与模糊表述。
生成符合《党政机关公文格式》（GB/T 9704—2012）规范、内容严守事实依据、问题直击要害、建议具可操作性的政策类报告；杜绝虚构政策、机构、案例、数据及未验证技术断言；所有文献引用须为2023–2024年正式发布或现行有效版本；输出为纯文本、零格式标记、Word直接可编辑格式。
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
将文本转换为超拟人语音。适用于"把这段文案读出来"等语音合成需求。
iFlytek Image Understanding (图片理解) — analyze and answer questions about images using Spark Vision model. WebSocket API, pure Python stdlib, no pip dependencies.
Recognize and extract structured data from invoices, receipts, and bills using iFlytek OCR API (科大讯飞票据识别). Supports VAT invoices, taxi receipts, train tickets, toll invoices, medical bills, bank receipts, and more.
ifly-pdf&image-ocr skill supporting both image OCR (AI-powered LLM OCR) and PDF document recognition. Use when user asks to OCR images, extract text from images/PDFs, convert PDF to Word/Markdown, or perform any OCR tasks on images or PDFs. Supports multi-language text extraction, document layout understanding, and various output formats.
Ultra-fast speech transcription service that converts audio files to text in record time - **1 hour of audio transcribes in ~20 seconds**.
iFlytek Official Document Proofreading (公文校对) — detect and correct errors in Chinese text including typos, punctuation, word order, factual mistakes, sensitive content, and more (27 error types). Supports up to 220,000 characters. Use when the user wants to proofread, check, or correct Chinese text, especially official documents. Pure Python stdlib, no pip dependencies.
iFlytek Machine Translation (机器翻译) — translate text between Chinese, English, Japanese, Korean, French, Spanish, German, Russian, Arabic, Thai, Vietnamese, and many more languages. Use when the user wants to translate text. Pure Python stdlib, no pip dependencies.
Translate videos with AI-powered dubbing using iFlytek (Xfei) Video Translation API. Supports video language translation, voice dubbing, multilingual video localization, 视频翻译, 视频配音, 视频本地化. Features task creation, task list, task details retrieval. Ideal for content creators and localization teams.
iFlytek Voice Clone tts(声音复刻) — train a custom voice model from audio samples and synthesize speech with the cloned voice. Supports the full workflow: get training text → create task → upload audio → submit training → poll results → synthesize with cloned voice. Pure Python stdlib, no pip dependencies.
PowerX STS 与插件鉴权规范（Exchange、KeyRing、拦截器、审计）。
PowerX gRPC 传输层规则（全局 server、拦截器链、DI）。
PowerX Nuxt API Client 规则。
>
在小程序项目里用 Vue SFC 写出“编译可预测、运行时可验证、类型可跟上”的页面和组件。
Builds mini-program UIs with Ant Design Mini components for Alipay and WeChat mini-programs. Covers Button, Form, List, Modal, Tabs, NavBar, and 60+ components with theme customization and CSS variable theming. Use when the user needs to create mini-program interfaces with Ant Design Mini, configure themes, or implement mini-program-specific UI patterns.
Builds enterprise React UIs with Ant Design (antd) including 60+ components (Button, Form, Table, Select, Modal, Message), design tokens, TypeScript support, and ConfigProvider theming. Use when the user needs to create React applications with Ant Design, build forms with validation, display data tables, or customize the Ant Design theme.
Add an ANSI color layer to existing ASCII/plain-text output (gradient/rainbow/highlights) with alignment-safe rules and a required no-color fallback. Use when the user wants to colorize terminal output, add rainbow effects to CLI text, or style ASCII art with ANSI colors.