Ficere

天机 —— 综合命理测算技能。一站式完成八字五行、称骨、紫微斗数（14主星+四化+大限）、西洋星座（太阳/月亮/上升）、三才五格姓名测算及多场景合盘；支持结构化 reading.json 输出与静态HTML可视化报告渲染。当用户提到命理测算、算命、八字、五行、称骨、紫微、星座、合盘、三才五格、姓名测算、姓名打分、名字吉凶、综合测算、一站式测算、天机、怎么和他沟通、如何安排工作、亲子养育建议时使用此技能。

降低AI生成内容(AIGC)检测率的文本改写技能。通过双阶段处理流程（润色+原创性增强）、 智能分段、上下文历史压缩、跨语言结构重塑等策略，将AI生成的学术论文或文章改写为 具有"人类写作印记"的文本。支持中文和英文，提供论文润色、学术增强、感情文章改写 三种模式。当用户提到降低AIGC检测率、降AI率、论文润色、学术改写、绕过AI检测、 人类化改写、anti-AI-detection、humanize text时使用此技能。

宏观推演：当用户需要宏观经济复盘、全球宏观情景推演、政策周期分析、大类资产敏感性、通胀/流动性/地缘事件分析、预期差识别、触发条件、失效信号和风险控制框架时使用。输出为非投资建议的结构化宏观研究。

Turn any article or copywriting into beautiful WeChat / Xiaohongshu image-card sets (图文包). Given article text and a cover image, inkflow generates: (1) a 16:9 AI cover illustration, (2) a 1080×1920 cover card with title + keyword highlights, (3) paginated body cards with clean typography, and (4) ready-to-publish copy with title alternatives. Outputs a ZIP of numbered PNGs plus optional Markdown copy. Supports 6 color themes: warm, cool_blue, sage_green, dusty_rose, amber, slate. Use when the user wants to turn copywriting, essays, or articles into scroll-friendly image cards for WeChat Official Account, Xiaohongshu, or any mobile-first platform.

Use when assessing a pharmaceutical pipeline asset, drug candidate, biotech license-in opportunity, BD asset, clinical-stage therapy, or investment feasibility brief. Produces reproducible MNC-style BD/License-in evaluation with ClinicalTrials.gov/PubMed evidence, 0-10 quantitative scoring, competitive landscape, risk flags, and Markdown plus JSON outputs.

Marsilea 可视化 Agent Skill：根据数据分析、真实数据案例或作图需求，生成可运行的 Python / Jupyter 示例， 覆盖注释热图、分组热图、聚类树、SizedHeatmap、CatHeatmap、layered heatmap、 ClusterBoard 布局、UpSet 图、Kaggle 数据集选型，并包含 uv/Jupyter 环境管理、中文字体处理、可复现实例和导出校验。 Marsilea visualization Agent Skill: turn data-analysis, real-world dataset, or plotting requests into runnable Python / Jupyter examples for annotated heatmaps, grouped heatmaps, dendrograms, SizedHeatmap, CatHeatmap, layered heatmaps, ClusterBoard layouts, UpSet plots, and Kaggle dataset selection, with uv/Jupyter setup, Chinese font handling, reproducible examples, and export checks.

Archer —— 面向特定领域专家的 AI 模型架构选型调研技能。在指定时间窗内系统性扫描 arXiv / bioRxiv / Hugging Face / 主流学术与中文技术源，筛选与领域相关的最新模型架构、预训练范式、tokenization 方案、长上下文与状态空间等通用 ML 创新，输出结构化的可关注架构清单（JSON 为主，Markdown 可选）。每条候选包含 title / summary / reason（为何对该领域有潜在价值）/ link 四要素。当用户提到模型架构调研、架构选型、追新架构、跟进最新模型、可关注架构、architecture survey、architecture watchlist、跟踪 arXiv、跟踪 bioRxiv、新架构筛选 时使用此技能。典型场景：基因组基础模型、蛋白语言模型、单细胞 FM、化学/材料 FM、视觉/多模态 FM、机器人 FM、RAG/Agent 架构跟进等领域专家的双周/月度选型调研。

Turns a user's dream narrative into a structured, non-fatalistic psychological interpretation. Extracts the narrative arc, emotional curve, symbols, and core conflicts, then generates multiple falsifiable interpretation hypotheses with confidence levels, physiological alternative explanations, clarifying questions, and light reflection actions. Supports pluggable interpretation frameworks (stress processing, psychodynamic, Jungian, cognitive script, continuity hypothesis, threat simulation, somatic, and an optional clearly-isolated traditional/cultural lens). Use when the user describes a dream and asks for 解梦 / dream interpretation / analyze this dream / what does this dream mean, or wants a dream journal entry structured into JSON/Markdown.

MarkCraft converts Markdown documents into polished, readable PDF reports with a technology-style visual theme. Use when the user asks to turn .md, Markdown, README, technical notes, product docs, research briefs, proposals, or long-form documentation into a PDF with readable typography, code blocks, left-aligned tables, citations, Chinese/CJK support, and professional layout. Optional team name and slogan can be added as decorative branding on the cover, header, and footer.

AI 技能全功能评测与管理工具。 【自动化模式】自动获取待评测技能、下载技能包、生成结构化评测并提交，支持状态持久化、限流退避、pending队列重试。 【人工评测模式】接收 skill_id 或技能链接，进行5-8组实测、生成完整评测报告。 【发布与管理】查询技能列表、上传/更新技能包、版本管理。

Biological experiment method (protocol/SOP) search, comparison, and structured retrieval for synthetic biology and drug discovery labs. Use when the user asks to: find an experimental method or SOP (LC-MS, fermentation, protein expression/purification, enzyme assay, molecular cloning, etc.), compare protocols, look up troubleshooting for a lab method, retrieve step-by-step parameters for a known SOP, search PubMed/Europe PMC/protocols.io for method literature, generate a grounded SOP draft, or check protocol completeness. Supports Chinese and English queries. Enforces compliance: internal full-text is returned; external sources return metadata + abstract + link only.

Uses knowledge graphs and adversarial multi-model reviews to enforce absolute numerical and logical consistency across interdependent Markdown and Excel files. Use when revising multi-file project proposals, financial models, or architecture docs.