digital-twin-creator · davepoon — команда для Claude Code
Команда digital-twin-creator проводит через сборку цифрового двойника: границы системы, аудит данных, калибровка по реальным замерам и прогон сценариев.
MIT ★ 3 785 обновлён 2026-10-25
/plugin marketplace add davepoon/buildwithclaude
/plugin install all-commands@buildwithclaudeЧто делает digital-twin-creator
Команда превращает запрос в пошаговый план цифрового двойника. Сначала выясняет, что именно моделируется и какие решения должен обслуживать двойник, потом описывает компоненты системы и то, что остаётся за её границами.
Дальше идёт работа с данными: каждый источник получает оценку от 1 до 10 по полноте, точности, актуальности, согласованности и релевантности, а пробелы сводятся в отдельный список. Затем команда задаёт каркас модели, калибровку и движок сценариев — со сверкой прогнозов по известным исходам и с живыми данными. Команда работает в Claude и Cursor.
Ваш текст уходит в шаблон как $ARGUMENTS, подсказка просит указать параметры двойника. Без ответов про объект, цель, нужную точность, доступные данные и частоту обновления модель строить не начнут: сначала пойдут уточняющие вопросы. Это честно — двойник без данных для калибровки остаётся красивой схемой.
Какие задачи решает
- очертить систему и её границы до начала моделирования
- оценить полноту и точность каждого источника данных
- сверить прогнозы модели с уже известными исходами
- настроить регулярное сравнение модели с живыми показателями
- проверить, как модель реагирует на смену параметров
Примеры запросов агенту
- Запусти
/digital-twin-creatorдля процесса обработки заказов: хочу проверить, что будет с пропускной способностью при другом потоке. /digital-twin-creator— нужен двойник клиентского пути, точность средняя, данные берём из выгрузки CRM.- Прогони
/digital-twin-creatorпо производственной линии и сразу сверь прогноз с фактическими замерами.
Как установить digital-twin-creator
1Через маркетплейс плагинов
/plugin marketplace add davepoon/buildwithclaude/plugin install all-commands@buildwithclaude2Вручную: положить файлы
.claude/commands/digital-twin-creator.mdВызов — /digital-twin-creator. Для всех проектов — ~/.claude/commands/
Вручную: положить файлы
~/.codex/prompts/digital-twin-creator.mdВызов — /prompts:digital-twin-creator. В Codex этот формат помечен устаревшим, замена — скиллы
Вручную: положить файлы
.github/prompts/digital-twin-creator.prompt.mdВручную: положить файлы
.gemini/commands/digital-twin-creator.tomlGemini хранит команды в TOML: текст запроса — в поле prompt
Вопросы о digital-twin-creator
Как установить digital-twin-creator в Claude Code?
Через маркетплейс: сначала /plugin marketplace add davepoon/buildwithclaude, потом /plugin install all-commands@buildwithclaude. Команда приезжает в составе плагина all-commands.
Что делает digital-twin-creator?
Ведёт от постановки задачи до калибровки: описывает компоненты и границы системы, оценивает качество данных, задаёт структуру модели, проверяет её на исторических исходах и готовит сценарии.
Что нужно указать при запуске?
Объект моделирования, решения, которые двойник должен поддерживать, требуемую точность, доступные реальные данные и частоту сверки с реальностью. Без этих ответов команда сначала задаст уточняющие вопросы.
Исходный текст
digital-twin-creator.md · MIT
--- description: Create systematic digital twins with data quality validation and real-world calibration loops. category: simulation-modeling argument-hint: "Specify digital twin parameters" --- # Digital Twin Creator Create systematic digital twins with data quality validation and real-world calibration loops. ## Instructions You are tasked with creating a comprehensive digital twin to simulate real-world systems, processes, or entities. Follow this systematic approach to build an accurate, calibrated model: **$ARGUMENTS** ### 1. Prerequisites Assessment **Critical Information Validation:** - **Twin Subject**: What specific system/process/entity are you modeling? - **Purpose & Decisions**: What decisions will this twin inform? - **Fidelity Level**: How accurate does the simulation need to be? - **Data Availability**: What real-world data can calibrate the model? - **Update Frequency**: How often will the twin sync with reality? **If any prerequisites are missing, guide the user:** ``` Missing Twin Subject: "I need clarity on what you're modeling. Are you creating a digital twin for: - Physical systems: Manufacturing line, vehicle performance, building operations - Business processes: Sales pipeline, customer journey, supply chain - Market dynamics: Customer segments, competitive landscape, demand patterns - Technical systems: Software performance, network behavior, user interactions" Missing Purpose Clarity: "What specific decisions will this digital twin help you make? - Optimization: Finding better configurations or strategies - Prediction: Forecasting future outcomes or behaviors - Risk Assessment: Understanding failure modes and vulnerabilities - Experimentation: Testing changes before real-world implementation - Monitoring: Detecting anomalies or performance degradation" Missing Fidelity Requirements: "How precise does your digital twin need to be? - High Fidelity (90%+ accuracy): Critical safety/financial decisions - Medium Fidelity (70-90% accuracy): Strategic planning and optimization - Low Fidelity (50-70% accuracy): Conceptual understanding and exploration" ``` ### 2. System Architecture Definition **Map the structure and boundaries of your target system:** #### System Components - Core elements and their relationships - Input/output interfaces and data flows - Control mechanisms and feedback loops - Performance metrics and success indicators - Failure modes and edge cases #### Boundary Definition - What's included vs. excluded from the model - External dependencies and influences - Environmental constraints and variables - Time horizons and operational contexts - Abstraction levels and detail granularity #### Relationship Mapping - Causal relationships between components - Correlation patterns and dependencies - Feedback loops and system dynamics - Emergent behaviors and non-linear effects - Lag times and temporal relationships **Quality Gate**: Validate that your system definition is: - Complete enough for the intended purpose - Bounded to avoid unnecessary complexity - Focused on factors that impact key decisions - Grounded in observable reality ### 3. Data Foundation Assessment **Evaluate and improve data quality systematically:** #### Data Inventory - Historical performance data and patterns - Real-time sensor/monitoring data streams - Configuration settings and parameters - External data sources and market conditions - Expert knowledge and domain insights #### Data Quality Analysis ``` For each data source, assess: - Completeness: What percentage of required data is available? - Accuracy: How reliable and error-free is the data? - Timeliness: How current and frequently updated is the data? - Consistency: Are there conflicts between data sources? - Relevance: How directly does this data impact key decisions? Quality Scoring (1-10 for each dimension): Data Source: [name] - Completeness: [score] - [explanation] - Accuracy: [score] - [explanation] - Timeliness: [score] - [explanation] - Consistency: [score] - [explanation] - Relevance: [score] - [explanation] Overall Quality Score: [average] ``` #### Data Gap Analysis - Critical missing information for model accuracy - Alternative data sources or proxies available - Data collection strategies for key gaps - Acceptable uncertainty levels for decisions ### 4. Model Construction Framework **Build the digital twin using systematic modeling approaches:** #### Component Modeling - Individual element behavior patterns - Performance characteristics and ranges - Response functions to different inputs - Degradation patterns and lifecycle factors - Optimization parameters and constraints #### System Interaction Modeling - Interface behaviors between components - Network effects and cascade influences - Resource sharing and competition dynamics - Communication protocols and latencies - Synchronization and coordination mechanisms #### Environmental Modeling - External factors affecting system performance - Market conditions and competitive dynamics - Regulatory constraints and compliance requirements - Economic factors and cost structures - Seasonal patterns and cyclical behaviors #### Dynamic Behavior Modeling - State transitions and evolutionary patterns - Learning and adaptation mechanisms - Scaling behaviors and capacity constraints - Stability and resilience characteristics - Performance under stress conditions ### 5. Calibration and Validation **Ensure model accuracy through systematic testing:** #### Historical Validation - Back-test model predictions against known outcomes - Identify systematic biases and correction factors - Validate model accuracy across different conditions - Test edge cases and extreme scenarios - Measure prediction error distributions #### Real-Time Calibration - Compare model outputs to live system data - Implement automated calibration adjustments - Monitor prediction accuracy over time - Detect model drift and degradation - Update parameters based on new observations #### Sensitivity Analysis - Test model response to parameter variations - Identify critical assumptions and dependencies - Understand uncertainty propagation through model - Validate robustness to data quality issues - Map confidence intervals for predictions **Calibration Metrics**: ``` Model Performance Dashboard: - Overall Accuracy: [percentage] ± [confidence interval] - Prediction Bias: [systematic error analysis] - Timing Accuracy: [lag prediction accuracy] - Extreme Event Prediction: [edge case performance] - Model Confidence: [uncertainty quantification] Recent Calibration Results: - Last Update: [timestamp] - Data Points Used: [count] - Accuracy Improvement: [change from previous] - Key Parameter Adjustments: [list] - Validation Test Results: [pass/fail with details] ``` ### 6. Scenario Simulation Engine **Enable comprehensive scenario testing:** #### Scenario Design Framework - Baseline/current state scenarios - Optimization scenarios testing improvements - Stress test scenarios with adverse conditions - What-if scenarios exploring alternatives - Innovation scenarios with new capabilities #### Simulation Execution - Automated scenario batch processing - Interactive scenario exploration interfaces - Real-time simulation monitoring and controls - Result aggregation and statistical analysis - Sensitivity testing across scenario parameters #### Output Generation - Performance metrics and KPI tracking - Visual simulation results and animations - Statistical analysis and confidence intervals - Comparative analysis across scenarios - Recommendation generation with rationale ### 7. Decision Integration **Connect simulation insights to actionable decisions:** #### Decision Framework Mapping - Link simulation outputs to specific decisions - Define decision criteria and thresholds - Map uncertainty levels to decision confidence - Establish risk tolerance for different choices - Create decision trees for complex scenarios #### Optimization Algorithms - Automated parameter optimization for goals - Multi-objective optimization with trade-offs - Constraint satisfaction for feasible solutions - Robust optimization under uncertainty - Dynamic optimization for changing conditions #### Recommendation Engine ``` Decision Recommendation Format: ## Scenario: [name and description] ### Recommended Action: [specific decision] ### Rationale: - Simulation Evidence: [key findings] - Performance Impact: [quantified benefits] - Risk Assessment: [potential downsides] - Confidence Level: [percentage with explanation] ### Implementation Guidance: - Immediate Actions: [specific steps] - Success Metrics: [measurable indicators] - Monitoring Plan: [ongoing validation approach] - Contingency Plans: [alternative actions if needed] ### Assumptions and Limitations: - Key Assumptions: [critical model assumptions] - Data Limitations: [known gaps or uncertainties] - Model Boundaries: [what's not included] - Update Requirements: [when to refresh model] ``` ### 8. Continuous Improvement Loop **Establish ongoing model enhancement:** #### Performance Monitoring - Automated accuracy tracking and alerting - Model drift detection and correction - Prediction error analysis and categorization - Data quality monitoring and improvement - User feedback collection and integration #### Model Evolution - Incremental model improvements based on learnings - New data integration and model expansion - Algorithm updates and enhancement - Scenario library expansion and refinement - User interface and experience improvements #### Learning Integration - Document insights from model successes and failures - Build institutional knowledge from simulation results - Share best practices across similar digital twins - Incorporate domain expert feedback and validation - Develop model confidence and reliability metrics ### 9. Output Generation **Present digital twin capabilities and insights:** ``` ## Digital Twin System: [Subject Name] ### System Overview - Purpose: [primary decision support goals] - Scope: [system boundaries and components] - Fidelity Level: [accuracy expectations] - Update Frequency: [refresh schedule] ### Model Architecture - Core Components: [key system elements] - Relationship Map: [interaction patterns] - Environmental Factors: [external influences] - Performance Metrics: [success indicators] ### Data Foundation - Primary Data Sources: [list with quality scores] - Data Quality Assessment: [overall quality rating] - Update Mechanisms: [how data stays current] - Validation Methods: [accuracy verification approaches] ### Simulation Capabilities - Scenario Types: [what can be modeled] - Time Horizons: [simulation time ranges] - Precision Levels: [accuracy expectations] - Output Formats: [reporting and visualization options] ### Calibration Status - Historical Validation: [back-testing results] - Real-Time Accuracy: [current performance metrics] - Last Calibration: [date and improvements] - Confidence Intervals: [uncertainty bounds] ### Decision Integration - Supported Decisions: [specific use cases] - Optimization Capabilities: [automatic improvement features] - Risk Assessment: [uncertainty and sensitivity analysis] - Recommendation Engine: [decision support features] ### Usage Guidelines - High Confidence Scenarios: [when to trust fully] - Medium Confidence Scenarios: [when to use with caution] - Low Confidence Scenarios: [when to gather more data] - Refresh Triggers: [when to update the model] ``` ### 10. Quality Assurance Framework **Ensure digital twin reliability and trustworthiness:** #### Validation Checklist - [ ] Model reproduces historical behavior accurately - [ ] Predictions are calibrated with confidence intervals - [ ] Edge cases and extreme scenarios are handled appropriately - [ ] Data quality meets requirements for intended decisions - [ ] Model boundaries are clearly defined and communicated - [ ] Assumptions are documented and regularly validated - [ ] Updates and maintenance procedures are established - [ ] User training and guidelines are comprehensive #### Risk Assessment - Model accuracy limitations and impact on decisions - Data dependency risks and mitigation strategies - Computational requirements and scalability constraints - User misinterpretation risks and training needs - System integration challenges and compatibility issues #### Success Metrics - Prediction accuracy improvement over time - Decision quality enhancement from model insights - Cost savings or performance improvements achieved - User adoption and satisfaction with digital twin - Model maintenance efficiency and cost effectiveness ## Usage Examples ```bash # Manufacturing optimization /simulation:digital-twin-creator Create digital twin of production line to optimize throughput and predict maintenance needs # Customer journey modeling /simulation:digital-twin-creator Build digital twin of customer acquisition funnel to test marketing strategies # Supply chain resilience /simulation:digital-twin-creator Model supply chain network to test disruption scenarios and optimization strategies # Software system performance /simulation:digital-twin-creator Create digital twin of microservices architecture to predict scaling and performance ``` ## Quality Indicators - **Green**: 85%+ historical accuracy, comprehensive data foundation, automated calibration - **Yellow**: 70-85% accuracy, good data coverage, manual calibration processes - **Red**: <70% accuracy, significant data gaps, limited validation ## Common Pitfalls to Avoid - Over-complexity: Modeling unnecessary details that don't impact decisions - Under-validation: Insufficient testing against real-world outcomes - Static thinking: Not updating model as reality changes - Data blindness: Ignoring data quality issues and biases - False precision: Claiming higher accuracy than data supports - Poor boundaries: Including too much or too little in model scope Transform your real-world challenges into a laboratory for exponential learning and optimization.
© BuildWithClaude Community · MIT · davepoon/buildwithclaude
Похожие слэш-команды
Все слэш-команды по теме «ИИ и ML» →Сначала — digital-twin-creator от других авторов.