shizaishiwo323

用于运行本地小红书图文自动化全流程：盘点已有素材，必要时爬取最新数据，把素材生成交给 xhs-material-pipeline，校验素材包，通过 DrissionPage 定时发布，并用 Gmail 汇报结果。

用于小红书图文素材包生成：分析素材、爆款参考、账号反馈和历史违规记录，产出标题、正文、封面/配图方案、发布 manifest、自动发布适配表，并完成图文包校验。

小红书图文素材生成、自动发布编排和相关本地流程共用的硬规则。用于去重 humanizer、imagegen、manifest、话题、历史违规、同主题去重和删除限制等跨 skill 约束。

Create narrated educational animation videos from knowledge documents and optional reference/model images using manimlib, edge-tts, and HyperFrames. Use when the user provides a Markdown/PDF/notes document and asks for a clear explainer animation, formula derivation video, scientific teaching video, voiceover, title/end cards, vector redraws of conceptual/model figures, dynamic visual explanations based on reference images, or validation that formulas and symbols inside ordinary text annotations render as real math instead of plain text.

Use when the user wants Codex to diagnose, plan, revise, or build prompts for AGU-style scientific manuscripts, with journal-specific adaptation for GRL, JGR, WRR, JGR: Solid Earth, and related geoscience journals. Especially useful for mechanism-driven geoscience, reactive transport, porous media, hydrogeophysics, or NMR papers. Applies teacher-derived OCAR/funnel structure, storyline alignment, narrative echo and payoff, methods evidence-chain, mechanism-centered results, paragraph engineering, sentence hierarchy, revision-case mining, and prompt-building rules. Trigger on requests about 前后呼应, 铺垫与兑现, Introduction-Results alignment, Results subsection transitions, or mechanism-driven AGU writing. Always diagnose manuscript stage and target journal profile before polishing sentences.

Use when a user needs real, high-relevance academic literature for a research topic, manuscript section, experiment, model, theory, review, method comparison, discussion, or reference-list strengthening.

Translate, polish, diagnose, and restructure SCI manuscripts into natural publication-ready English while preserving scientific meaning and evidence. Use for Chinese-to-English scientific translation, sentence or paragraph polishing, logic and cohesion repair, IMRaD section revision, abstract/title editing, full-manuscript audits, and removal of AI-like invented compound labels. Preserve standard technical hyphenation and never invent results, citations, mechanisms, or unsupported claims.

Polish academic manuscript sentences while preserving scientific meaning, evidence strength, citations, variables, equations, figure labels, paragraph order, and manuscript logic. Use when the user asks to reduce AI-like academic wording, simplify paper sentences, define unclear terms, remove unsupported overclaiming, handle hyphenated or dash-based compounds, or make scientific prose clearer, more restrained, and more natural without changing the paper's content or storyline.

Convert a provided video plus subtitle file into Chinese Markdown notes with extracted screenshots inserted at useful visual reference points. Use when the user gives video/subtitle artifacts such as .mp4 with .srt, .vtt, .ass, or .ssa subtitles and asks for Markdown notes, lecture notes, transcript notes, screenshot placeholders, screenshot insertion, or subtitle-to-notes conversion while preserving all subtitle text.

Use when building or extending a scientific computing AI agent that wraps existing domain code with guided parameter selection, tool calling, Streamlit UI, result interpretation, reports, artifacts, or deployment; especially for NMR, geophysics, inversion, simulation, meshing, preprocessing, and analysis workflows.