Citrus-bit

Use this skill for academic reports, literature reviews, experiment reports, user-selected reference styles, innovation-point mining, or when the user wants a topic turned into a paper-backed evidence bundle instead of generic web research.

Use this skill whenever the user wants reproducible CS/AI experiments, model evaluation, regression/classification/clustering analyses, bioinformatics workflows, QC, differential expression, single-cell starter analysis, or manuscript-ready experiment bundles…

Generate a personalized SOUL.md through a warm, adaptive onboarding conversation. Trigger when the user wants to create, set up, or initialize their AI partner's identity, or when a SOUL.md is missing. Also trigger for updates to an existing SOUL.md or AI…

Interact with MedrixFlow AI agent platform via its HTTP API. Use this skill when the user wants to send messages or questions to MedrixFlow for research/analysis, start a MedrixFlow conversation thread, check MedrixFlow status or health, list available…

Use when the user asks for latest datasets, benchmark suites, leaderboards, baseline or SOTA comparisons, dataset selection for experiments, or evidence that a paper needs complete experimental validation. This skill routes to the dataset_benchmark_discovery…

Use this skill for general web research that needs current online information, multiple source angles, and synthesis, when no more specific research skill applies. Prefer academic-deep-research for scholarly literature, github-deep-research for…

Use when the user wants to create any technical diagram - architecture, data flow, flowchart, sequence, agent/memory, or concept map - and export as SVG+PNG. Trigger on: "画图" "帮我画" "生成图" "做个图" "架构图" "流程图" "可视化一下" "出图" "generate diagram" "draw diagram"…

Create distinctive, production-grade frontend interfaces with high design quality for new standalone web artifacts, pages, posters, dashboards, React components, HTML/CSS layouts, or clearly requested creative UI builds. For modifications inside an existing…

设计、实现或评审以简历证据为基础的模拟面试流程，包括受控题库检索、岗位化面试计划、动态追问、STAR 故事卡、语音转写确认、回答评分与辅导，以及回答和简历口径一致性检查。用于新增或改进 question.retrieve、interview.plan、story.build、speech.transcribe、answer.evaluate、answer.coach、resumeInterview.check，或处理“生成面试题”“模拟面试”“评审录音回答”“检查简历口径”等请求。

审计、实现或回归验证简历助手及其 Skill 的安全和质量边界，包括 PII 脱敏、提示注入防护、Capability 权限、密钥和日志隔离、文档 worker 沙箱、无障碍、依赖与许可证供应链，以及 LLM 结构化输出和事实安全评测。用于接入或评审 pii.redact、prompt.guard、accessibility.audit、security.audit、llm.eval，新供应商、新 Skill、发布门禁、隐私检查或安全回归请求。

Unified entry point for planning, implementing, reviewing, or extending the Resume Analysis Assistant across PDF parsing, evidence-backed resume analysis, JD matching, professional writing, Typst export, interview coaching, privacy, security, accessibility,…

基于 Resume AST、SourceBlock 和 EvidenceGraph 分析、实现或评审简历证据链、声明支持状态、事实冲突、六维评分、持续 AI 编辑对话、安全修改建议与 ATS 机器可读性审计。处理证据挖掘、数字或经历真实性、可追溯扣分、对话上下文、needs_proof/ask_user 流程、JSON Patch/revision 安全，或修改…

分析、实现、诊断或评审简历 PDF 的安全接入、原生文字与坐标提取、扫描/混合页局部 OCR、块去重、阅读顺序恢复、语义分段和 Resume AST 来源关联。处理 PDF 上传、乱码或无文字层、双栏/表格版面、解析 warning、worker 与 PDF.js 回退，或修改 document.parse、document.ocr、document.segment Capability 时使用。

Analyze job descriptions and recruitment risks, build a complete JD requirement-to-evidence-to-gap-to-recommendation matrix, tailor resume emphasis, and rewrite Chinese or English career copy without changing facts. Use when Codex must parse or assess a JD,…

Recommend, render, preview, audit, and export ATS-friendly resume PDFs through three Typst templates with strict content, font, visual, integrity, and user-confirmation gates. Use when Codex must choose or refine resume layout, generate…

一个全能型产品经理助手，擅长运用第一性原理思考，帮助用户从零构建产品需求。它能够生成详细的 PRD、Changelog 和 System Prompt，并熟练调用 AI 工具（如 Cursor/Claude, Google AI Studio）进行辅助设计。适用于产品构思、需求定义、文档维护等场景。