在Backtrader中实现多支股票的回测，并通过设置数据源的_name属性，在策略的stop方法中区分并输出对应股票的信息。
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 →在Backtrader中实现多支股票的回测，并通过设置数据源的_name属性，在策略的stop方法中区分并输出对应股票的信息。
Translates English text into Chinese verses using a modern poetic style, adhering to specific character counts if requested, while prioritizing elegance, rhythm, and imagery over literal translation.
使用Java和Jsoup库解析特定结构的HTML赔率表格，提取赔率公司、水、盘、水以及变化时间。
将长周期的K线数据按照OHLC聚合规则压缩为指定长度的K线序列，并进行归一化处理。
解析包含教育经历的JSON字符串，并根据<TOKEN>和education_level字段对列表进行排序，支持正序和倒序。
针对用户指定的Python库（如Kafka, Elasticsearch, RabbitMQ），生成包含数据获取、存储及特定接口（如创建索引）的类封装代码。
编写Python脚本，使用正则表达式从特定格式的婴儿护理日志文本中提取日期、时间和配方奶量，并输出为JSON数组。
针对TensorFlow训练代码进行内存泄漏修复，包括优化数据管道、添加每轮结束后的垃圾回收回调以及修正ModelCheckpoint配置。
Translates Simplified Chinese text from a Pokémon novel into English, ensuring the meaning is conveyed and comprehensible to non-Chinese speakers, while strictly using official Pokémon franchise names for characters and species.
根据提供的对话或材料，严格按照五段式结构撰写议论文或提纲。结构要求包括：第一段明确提出论题，第二段说明论题，第三、四段提出分论点并举例论证，第五段总结全文。
使用Python的ThreadPoolExecutor将串行的股票数据下载任务改为并发执行，并利用tqdm进度条实时展示当前处理的股票代码。
将德语学术文本（特别是哲学、政治经济学和社会理论领域）准确翻译为中文，保持原文结构、术语准确性和引用格式。
根据小说或场景描述，生成符合特定JSON格式的视频分镜提示词。严格保持场景背景、光照、角度等静态元素不变，仅根据剧情变化人物动作和表情。
将英文医学、生物信息学及学术文本翻译为符合中文期刊发表规范的专业学术语言。
根据提供的文章生成英语主旨大意或标题选择题，要求选项单词数限制在7词以内，结构对仗工整，并利用扩缩范围、无中生有等策略设置干扰项。
根据产品信息生成适用于展销会或路演的6分钟双人英语演讲稿，内容严格分为自身特色/目标客户和市场对比/竞争优势两部分，并附带详细的时间分配表。
Extracts specific structural components—Project Topic, Overarching Research Questions, Identified Method (with literature), Sampling Strategy, Data to be Collected, and Method Planning—from a provided academic research proposal text.
Writes the 'Empirical Analysis Results' section for an undergraduate economics thesis, adapting provided reference text to the specific topic of media impact on stock market volatility and stability while ensuring low plagiarism and academic credibility.
Translates Simplified Chinese text from a Pokémon novel into English, ensuring clarity for non-Chinese speakers while strictly applying official Pokémon franchise terminology and character names.
Extracts and formats specific details (coupon ID, time, duration) from promotional activity descriptions with strict formatting rules.
General SOP for common requests related to gr, gradio, file.
专注于Gradio及Hugging Face空间的Dockerfile配置与优化顾问，基于Python 3.10和阿里云镜像源生成或修改Dockerfile，解决部署问题。
将中文室内设计场景描述、关键词及建筑特征翻译成适用于Midjourney的英文提示词。
General SOP for common requests related to motorcontrol, temp, motor.
General SOP for common requests related to node, value, binarytree.
General SOP for common requests related to os, asyncio, path.
Guides the user on adding a new TEXT column to a PostgreSQL table using SQLx migrations and backfilling existing rows with nanoid values, specifically targeting rows where the column is NULL.
General SOP for extracting and structuring processes or checklists from conversation evidence.
SOP for PyTorch setup involving specific input splitting logic and standard library imports.
General SOP for common requests related to self, current, if.
General SOP for common requests related to sugar, can, mg.
General SOP for generating or analyzing training logs, incorporating metrics like loss, accuracy, precision, recall, and max accuracy across epochs.
用于在Unity中配置和实现ML-Agents卡牌游戏智能体，包括定义包含手牌、法力、生命值、场上随从及牌库信息的观察空间，以及设置离散动作空间和回合重置逻辑。
Specializes in creating TikTok short video scripts using an anthropomorphic 'Chef' perspective to explain wildlife food chains and survival environments, utilizing real field footage.
General SOP for common requests related to 亲爱的, 天才俱乐部, 81.
专门处理俄语文本，提供自然流畅、优美高雅的中文翻译或专业总结，适用于教育、行政及新闻等场景。
General SOP for common requests related to 六四事件, 1989, 然而.
根据提供的文本资料，生成包含教育体系、机构数量、师资力量及政府政策四个模块的学术性区域教育发展报告。
General SOP for common requests related to 商业模式, 期prp项目, 产品服务.
将学术或正式文本改写为符合公众号风格的文案，旨在吸引读者阅读兴趣，语言更生动活泼。
用于撰写关于国内科研设施（如HIAF）的学术论文引言。核心策略是通过对比国际成熟系统与国内发展现状来确立研究动机，规范引用的国际同类设施，并约束学术用词的正式性。
根据用户提供的求职信模板和目标职位，仿写一篇包含段落结构标签、相关资格证书及附件信息的求职信。
去除新闻文本中的主观情绪、修饰性定语和心理暗示成分，仅保留客观事实和数据，以提取新闻的核心价值。
根据提供的对话内容（通常是科学家与文学家的对话），按照特定的五段式结构要求撰写议论文或提纲。
General SOP for common requests related to 身边, 心里, 脸上.
Executes a 1-back working memory test protocol by comparing the current input in a sequence to the previous one, stating the prior item and indicating a match or mismatch.
Generates 2-paragraph informational or narrative texts for 5th graders designed to practice inference and main idea identification. The output must include 5 questions about main/supporting points and 2 inference questions, aligned with specific ELA learning objectives.
Identifies the critical 20% of a topic that yields 80% of the desired results and creates a focused learning plan to master it.
Shortens academic abstracts to a strict maximum of 160 words while adhering to a specific four-part structure: problem statement, background context, primary results, and a concluding sentence.
Generates academic articles with a specific structure (Intro, 4 body paragraphs, Conclusion), academic tone, extensive in-text citations, and a strictly formatted APA-7 reference list including a signature block.