OliviaL
Weekend Family Outing Planner
Not sure where to take the kids this weekend? Get one clear recommendation. It remembers your family’s details: your city, each child’s age, budget range, whether you drive or take public transit, places you want to avoid, and even which parking lots not to enter. After asking once, it won’t ask again—you can simply say, “Where should we go this week?” It doesn’t guess the weather. Each time, it checks the actual forecast for your target date online. It won’t schedule outdoor activities for a 33°C afternoon, prioritizes indoor options on rainy days with a backup plan, and even factors sunset time into scenic outings. It only recommends places that are genuinely open and available: primarily using Xiaohongshu and Douyin content from the past 1–3 months, covering your local area and destinations within a 2-hour drive. Posts older than 3 months are not used as evidence that a place is currently visitable. Events are verified to still be running this week, and venues are verified to still be open. Every plan includes the link to the actual post found during research—URLs are never made up. It counts adults as one person and evaluates each option across six dimensions: budget, weather, transportation, age suitability, educational and physical activity value, and whether both adults and children can enjoy it. You’ll get 1 top recommendation plus 1–2 alternatives with meaningful differences. “Adults should have something fun to do too—not just supervise playtime” is a required scoring criterion. Once you choose a plan, it will ask whether you’d like an itinerary card: an HTML webpage that works on your phone, supports check-off tracking, and offers one-tap navigation. Or, it can generate a vertical image card from the webpage card so you can save it to your photo library with a long press and share it directly in your family group chat—even without a signal. The time, location, and price on the image card are copied from the webpage card line by line, without adding a single word. The more you use it, the easier it gets—preferences confirmed in each round are saved to your family profile.
Positioning for Profit
You don’t lack capabilities—you lack a positioning statement you can sell. “Positioning for Profit” uses eight questions to turn scattered experience and resources into an evidence-backed business positioning statement: a two-step method identifies the money-making quadrant you should be in right now, draws out your advantage sweet spot and commercial strengths from your real experience, then uses the Golden Triangle to verify whether how you earn, why you can, and whose money you earn are aligned. No labels, no guesswork, and no mistaking interest for an advantage. After passing seven quality gates, you’ll receive a positioning card, decision card, positioning report, and product-planning handoff package. Every unverified conclusion is tied to a falsifiable action. Positioning is a current choice—not a lifelong label.
TRIZ Research Innovation Coach
TRIZ Research Innovation Coach is a system for discovering innovative research problems and generating research topics for academic work. Centered on TRIZ contradiction resolution, the nine-screen method, laws of evolution, resource analysis, and thinking-trap diagnosis, it uses a rigorous AFP-style step-by-step process to transform vague research challenges into clear scientific questions, actionable research plans, and theoretically compelling topic directions. The system also includes dual review by a thinking coach and an academic editor, assessing research subjects, theoretical mechanisms, evidence pathways, methodological fit, and degree of innovation. It helps users overcome common challenges such as “I can’t find an innovation,” “I can only compromise,” and “my methods are limiting me,” while building transferable and reusable innovative thinking skills.
Skill Quality Audit (PDCA-QMS)
Evaluate Skills using a quality management system based on management science, not an off-the-cuff checklist. Based on Six Sigma CTQ/FMEA/DPMO × TQM × Deming PDSA cycle, through the Plan-Do-Check-Act four phases, it provides: a quality scorecard with Sigma level, a defect priority table ranked by RPN, root cause analysis down to the instruction design layer, and the full Skill text after Poka-Yoke correction. The fixer and reviewer are forcibly separated, and scores only increase, never decrease. v2.0 additions: ① Quantitative contract – closed enumeration of structural unit counting (only phase level), K-value gradient anti-scoring lock, Sigma table lookup with log axis interpolation, and full zero-padding to prevent division by zero, ensuring scores are reproducible and undistorted; ② Quality level × disposal path dual-axis gate, eliminating the contradiction of 'unqualified but recommended for release' labels; ③ Diagnostic mode – say 'only a report' or 'don't modify yet', then run through P-D-C and stop at Check, without modifying your script; ④ Output contract – pass/fail points fold, evidence limited to 50 characters, phase word count limits, values must include calculation process. When to use: evaluate whether a Skill is good, diagnose why its output is unstable, systematically optimize an existing Skill, benchmark multiple Skills, pre-release quality inspection of Skill drafts, or only want a diagnostic report without modifying the draft. Trigger words: detect skill quality, skill quality check, quality audit, rate a skill, PDCA optimize skill, skill health check, evaluate this skill, skill defect analysis, diagnose only without repair, pre-release skill quality check, skill benchmarking.
Parent-Child Treasure Hunt
Generate ready-to-print, ready-to-run parent-child treasure hunt puzzle games for parents of children ages 5–10: child-only clue cards (clues only, picture-book-quality visuals and text, no parent content) plus a separate parent answer sheet for setup, answers, hints, and role-play. The child’s personal information is remembered automatically.
Critical Writing Pipeline
From a vague idea to a rigorous argumentative essay with verifiable evidence chains — a full-pipeline AI writing system supporting both Chinese and English. Optimized with AFP 3.1: five-step nested architecture, Core B 9-dimension quantified audit, adaptive pacing, ratchet version management, cross-session state persistence, and material verification downgrade strategy.
TRIZ-Academic Problem Discovery v2.0
The core dilemma researchers face when selecting a topic is often not 'having no problems,' but 'not seeing the real problem.' TRIZ-Academic Problem Discovery v2.0 uses the AFP 3.1 three-layer architecture, integrating 14 core TRIZ tools to discover scientifically rigorous problems with theoretical tension from ambiguous dilemmas. Core Capabilities: 1. 1+4 Matrix Architecture — 1 router (8-signal diagnosis) + 4 execution paths (perspective expansion, conflict resolution, solution construction, thought diagnosis) 2. Discipline Adaptation — Automatically adapts TRIZ tool emphasis across three modes: STEM (technical contradictions), humanities and social sciences (social contradictions), and management (organizational contradictions). 3. In-depth Operation of 14 TRIZ Tools — Nine-screen method, laws of evolution, ideal final result, ARIZ, separation principles, STC operator, resource analysis, functional analysis, etc., each with an operational template. 4. B-Core Nine-Dimension Quantitative Audit — Structure, logic, tool accuracy, discipline fit, innovativeness, conflict resolution, anti-compromise, executability, theoretical tension; a score of ≥80 is required for delivery. 5. Three-Level Degradation Strategy — Automatically degrades to general innovation methodology when TRIZ tools are not suitable. Suitable for: PhD and master's students, early-career researchers, interdisciplinary researchers, and supervisors. Supports both Chinese and English.
Auspicious Posting Time
Based on the user's Ba Zi (or automatically converted from birth information) and the current date, it fetches the day's Huang Li online. It then scores from four dimensions: the relationship between the day master and the flowing day (including Xiyongshen correction), Huang Li's auspicious and inauspicious activities, gods and spirits' fortune, and timing suitability. After being audited by the B-core fate quality check gate, it generates a posting fortune report. Supports Ganzhi verification, data degradation, and cross-conversation state persistence.
All-in-One Review Response V3.0
A full-process system for responding to peer review comments in social sciences and economics/management academic papers — from comment analysis, strategy formulation, and hard flaw defense to one-click assembly of a Rebuttal Letter. V3.0 Core Upgrades: · Governance layer: unified dashboard + 3 step modes (Standard/Fast/Slow) + ratchet version management + mid-process join logic · All 12 quick commands implemented (/quick /jump /skip /rerun /export /resume /tone /hardcase /conflict /cover /draft /lang) · Cross-step comment tracking table: tracks the status of each comment from Step 1 to Step 6 for closed-loop management · Complete library of 7 hard flaw diagnostics + subfield adaptation (Accounting/Finance/Management/Marketing/Labor Economics/Macroeconomics/Qualitative Research) · Full language pair support: Chinese→English / English→English / Chinese→Chinese, with automatic detection of review comment language · New revision Cover Letter generation + summary table of changes · Tone calibration system upgrade: forbidden expression list (20 items) + auto-correction + politeness score · Enhanced safety boundaries: anti-hallucination citation check + anti-overpromising check + AI-generated content transparency statement Applicable scenario: One-stop response writing for PhD students and early-career researchers in social sciences and management after receiving review comments.
Journal Pre-review Expert V7.0
Top Journal Submission Pre-review Expert V7.0 is a full-process pre-review Skill for social sciences and humanities scholars. It simulates the perspective of senior reviewers at Nature/Science/ASR/AJS-level journals, driven by a dual-core engine (Reviewer Butcher + Polishing Craftsman) across six phases. V7.0 core upgrades: · Governance layer: unified dashboard + step mode (default/fast-forward/slow-motion) + ratchet version management + mid-join logic · All 15 quick commands implemented (/express /jump /benchmark /ethics-scan /cover-letter /polish /abstract /checklist /audit /resume /skip /rerun /export /persona /lang) · Cross-phase vulnerability tracking table: tracks from Phase 2 to Phase 5 with status closure · 6 virtual reviewer personas (methodology hawk, theoretical purist, contextualist, ethics reviewer, etc.), each with 4-5 typical criticisms, automatically activating 3-4 based on research type · Full language pair support: Chinese→English, English→English, Chinese→Chinese, with Chinglish terminator engine covering 7 categories of high-frequency errors · Benchmark paper five-dimensional comparison + no-benchmark degradation path · New AI usage transparency check and citation quality deep audit Applicable scenarios: PhD students/early-career faculty pre-submission self-check, advisors guiding students to revise manuscripts, research group submission strategy discussions. Supports full-process guidance for Chinese native speakers submitting in English.
ImageMy Baby is the Storybook Hero
Make your child the star of a storybook! Just upload a photo, and AI generates a personalized storybook, blending your child's real face into beautiful story scenes. Supports all ages 0–12, automatically adjusting story complexity, sentence length, and art style suggestions based on the child's age. Three character design styles (fully realistic, realistic face + cartoon body, fully cartoon), five art style choices, with 100% character consistency locking. Output is a 3:4 vertical picture storybook suitable for Xiaohongshu/WeChat Official Accounts, with an optional promotional poster. ✨ Key Advantages 1. Semi-realistic animation style (Disney live-action feel) — cinematic visual quality, perfectly blending real facial features with animated scenes. 2. Character consistency engine — automatically verifies facial features, hairstyle, and clothing consistency page by page, regenerating if unqualified. 3. Age adaptation — four tiers (0–3, 3–6, 6–9, 9–12) automatically adapt story depth, language style, and visual suggestions. 4. One-stop creation — photo → script → images → verification → delivery → poster, a full-process closed loop.
SocialScienceCompletionV7.0
A provincial/department-level social science project completion report generation system based on the AFP 3.1 three-layer architecture. New additions: three-layer nested governance (framework/content separation) + dual-core parameterization (R1-R6 role library) + 9-dimensional quantitative scoring matrix + ratchet version management + cross-conversation persistence + defect P0-P3 classification. Input the project proposal + published papers, complete a 20,000-50,000 word comprehensive report and 13 deliverables in 2-3 days. 6 iterations accumulated experience in handling 13 failure modes.
ResearchBuild Yourself as a Nation v4.0
Plan your personal growth using the strategic logic of nation-building. Through questionnaire-style guidance and scaled assessments, this Skill helps you systematically build yourself like a chief architect—from a “Founding Ceremony” to a “State of the Nation Address.” Set a lifelong vision, anchor milestones, write a five-year plan, create annual plans, break them down into quarterly and monthly actions, balance three domains, use subtraction-based closure, and run national-defense stress tests while continuously reviewing and adjusting your course. Once built, keep it running: update it monthly and manage yourself like a nation. Trigger words: nation-building / five-year plan / life planning / annual plan / personal growth / goal management / State of the Nation Address.
ResearchMgmt C-J Intr Expert v3.0
Writing the introduction for management C-journal papers is a challenging task for many researchers. This Skill v3.0 adopts the AFP 3.1 three-layer architecture, providing a closed-loop process from information collection to final delivery. Core capabilities: 1. Anti-hallucination engine—mandatory input + traceable verification: all references must come from the user's provided pool of real literature; the AI cannot supplement or fabricate. 2. Dual-dimension adaptation (discipline × research type)—four disciplines: strategic management, organizational behavior and human resources, public management, and information management, × three types: empirical quantitative, qualitative case, and theoretical construction. Automatically matches 12 introduction paradigms. 3. Precise journal style adaptation—uses googleSearch to retrieve examples from target journals in the last 3 years, with a three-level fallback strategy to ensure retrieval success. 4. Five-part introduction structure theory—research background → transition → research object → research value → research design, generating a blueprint paragraph by paragraph. 5. B-core nine-dimension quality audit—structure, logic, citation, adaptation, style, innovation, gap, theory, anti-hallucination. Delivery only if quantitative score ≥80. Suitable for: graduate students and early-career researchers in management who need to submit to CSSCI/Beijing University Core journals.
ImageHealing Rant Comic Generator
Turn your rants into adorable storybook comics. Pick from 12 classic characters or upload your own. The AI automatically creates a two-panel 'Expectation vs Reality' script and generates 2-head-tall chibi comics. Confirm at each step to maintain character consistency and stylistic unity. Use cases: workplace venting, life's little frustrations, product experiences, social dilemmas.
SlidesThesis Defense: PPT*Script*QA
Say goodbye to defense anxiety! Generate academically standard PPT and word-for-word scripts from a 100,000-word thesis with one click. Each slide is strictly synchronized with the script, and timing is precisely controlled. Additionally, an AI defense committee provides mock Q&A (three difficulty levels: easy, medium, hard) that deeply questions weaknesses in your thesis and offers professional feedback. Adaptive to four major disciplines: Science/Engineering, Humanities, Social Sciences, and Arts, covering proposal, mid-term, and graduation defenses. Suitable for master's and doctoral defenses and proposal reports.
ResearchYM-Skill Competitor Analyst V3.0
YouMind Skill Marketplace professional-grade competitive analysis system. Supports three input methods: skill ID, keywords, and requirement description. Deep analysis from six dimensions: market overview, feature comparison, credit pricing, differentiation, market gaps, and competitive strategy. Built-in 8-dimensional quantitative audit matrix + dual-core engine ensures data authenticity and analysis quality. Supports adaptive stepping and iterative optimization. Generates actionable competitive analysis reports and visual infographics in 3 minutes.
WeChat Article Creator v3.0
WeChat Official Account valuable content writing engine based on the AFP 3.1 three-layer architecture. It enforces real-life cases, a three-layer de-AI flavor defense, and B-core nine-dimensional adversarial quality checks. Delivers a complete content package including the main article, cover image, and inline images. Features a 7-step process with adaptive progression. Suitable for: self-media valuable articles, long-form tool reviews, and in-depth methodology shares. Trigger words: write WeChat official account valuable content, AFP structure article, official account long article, de-AI flavor writing.
Academic Lens: Deep Paper Notes
Get to the essence of papers from a critical perspective. Generate deep notes that can be directly imported into Obsidian in 15 minutes—not just summaries, but revealing theoretical dialogues, causal mechanisms, and research gaps. Supports full-discipline adaptation, three-level reading modes, intelligent routing by paper type, and hallucination protection.
Public Admin Grant Assistant v2.0
Intelligent writing system for Ministry of Education Humanities and Social Sciences Fund project applications (Public Administration focus). Built-in 8-dimensional quantitative audit matrix + dual-core engine (A-core construction / B-core independent scoring), covering full-process quality control from topic selection diagnosis, web-wide search, literature review, research design to outcome planning, supporting fast-forward/slow-play adaptive stepping. Built-in logical chain review and policy-strategy alignment verification, one-click generation of high-quality applications and full supporting assets compliant with review standards.
University Faculty IP Planner
An IP positioning tool designed specifically for university faculty—using a six-dimensional academic asset canvas to precisely translate academic accumulation into IP directions, with built-in compliance boundary reminders and academic year rhythm adaptation, ensuring that teachers' academic assets are no longer confined to papers.
WriteSmart Classroom Design v4.0 B
Transform vague teaching ideas into well-structured smart classroom plans. 6-stage guided design with dual-core quantitative quality checks ensures alignment of objectives, activities, and assessments. Three-mode adaptive progression supports all grade levels and subjects. Simply say 'Help me design a lesson' to get started.
Doctoral Thesis Advisor v2.0
A full-process thesis writing guidance system driven by three thinking models: Big Loop, Iron Triangle, and Russian Doll. From outline design to body writing to delivery and compilation, each step enforces academic structural compliance checks to ensure rigorous logic and complete argumentation. Suitable for doctoral, master's, and undergraduate thesis writing.