Draft and revise Chinese technical articles and notes with clear conclusions and supporting evidence. Preserve the author's experience, important quotations, and technical detail while distinguishing facts from inferences and improving the structure where needed.
Write or refine Chinese technical notes and articles in a concise, evidence-led voice. Use for substantive drafting or rewriting, not every Chinese reply.
Review the instructions the agent will receive before you install.
先确认读者、用途和落点。用户指定的语气与结构优先；有现成文章时只读当前项目中最相关的两三篇，不为模仿文风搜索其他工作区或私有历史。
技术方案按问题、机制、取舍、证据、下一步组织，短修改不套整篇模板。面向大众时先解释具体情境，再介绍术语；手机长图或视频仅在用户需要时选择当前可用工具，不依赖某台机器的渲染器。
修改已有协作文档，先读最新内容与评论、高亮等人工标记，再做局部修改并读回；不要用整篇覆盖抹掉协作线索。同步或发布按当前任务授权执行，写稿本身不授权发送。
持续维护飞书文档或同步本地 mirror 时，使用 managed-lark-docs 的局部 patch、协作标记保护与读回流程；文风要求不扩大远端写入范围。
完成时检查：读者能否理解核心判断、证据是否支持它、重要原意是否保留、引用是否可追溯、是否混入私有来源或编辑过程。
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Gather recent AI products, papers, tools, and viewpoints into a Chinese and English HTML daily report. Select up to ten supported items, check original sources where possible, and summarize shared trends with practical next actions.
Retrieve an arxiv paper's full text from its HTML version using an ID or URL. Extract readable text with section structure, optionally save it to a file, and prepare it for reading, notes, or technical analysis.
Search the web or find images through the Ask-Echo API from 火山引擎. Retrieve result details, request an AI-generated summary, and narrow searches by time range or selected websites.
Build or update the BlueBubbles channel plugin for Clawdbot. The skill explains code locations, existing helper functions, message delivery, connection checks, and webhook handling, which receives incoming messages and attachments through HTTP notifications.
Turn API documentation, SDKs, or existing scripts into reusable command-line tools for Codex. Build focused read and write commands, stable JSON output, authentication checks, and a companion skill explaining how future sessions should use the tool.
Improve existing code through up to two optimization rounds, measuring execution time and memory use against a baseline. Verify that results remain correct, save the optimized code, and document changes, measured gains, and trade-offs in a report.