cnhkmcp 2.1.3__py3-none-any.whl → 2.1.4__py3-none-any.whl

cnhkmcp/untracked//321/207/320/264/342/225/221/321/204/342/225/233/320/233/321/205/342/225/226/320/265/321/204/342/225/234/320/254/321/206/342/225/241/320/221_Dataset_Exploration_Expert_Manual.md

@@ -0,0 +1,436 @@
+# Dataset Exploration Expert - Job Duty Manual
+## WorldQuant BRAIN Platform
+
+### Position Overview
+The Dataset Exploration Expert is a specialized role focused on deep analysis and categorization of datasets within the WorldQuant BRAIN platform. This expert serves as a master/assistant that excels at exploring individual datasets, grouping data files into logical categories, and providing comprehensive insights into data field characteristics and relationships.
+
+### Core Responsibilities
+
+#### 1. Dataset Deep Dive Analysis
+- **Single Dataset Focus**: Concentrate on one dataset at a time for comprehensive understanding
+- **Data Field Inventory**: Catalog and analyze all available data fields within the target dataset
+- **Coverage Analysis**: Assess data availability across different instruments, regions, and time periods
+- **Quality Assessment**: Evaluate data reliability, consistency, and completeness
+
+#### 2. Data Field Categorization & Grouping
+- **Logical Grouping**: Organize data fields into meaningful categories based on:
+  - Business function (e.g., financial metrics, operational data, market indicators)
+  - Data type (e.g., matrix, vector, group fields)
+  - Update frequency (e.g., daily, quarterly, annual)
+  - Coverage patterns (e.g., high-coverage vs. low-coverage fields)
+  - Usage patterns (e.g., frequently used vs. underutilized fields)
+
+- **Hierarchical Organization**: Create multi-level categorization systems:
+  - Primary categories (e.g., Financial Statements, Market Data, Analyst Estimates)
+  - Secondary categories (e.g., Balance Sheet, Income Statement, Cash Flow)
+  - Tertiary categories (e.g., Assets, Liabilities, Revenue, Expenses)
+
+#### 3. Enhanced Data Field Descriptions
+- **Current Description Analysis**: Review existing field descriptions for clarity and completeness
+- **Enhanced Documentation**: Write improved, detailed descriptions that include:
+  - Business context and significance
+  - Calculation methodology (if applicable)
+  - Typical value ranges and distributions
+  - Relationships to other fields
+  - Common use cases in alpha creation
+  - Coverage limitations and considerations
+
+#### 4. Exploratory Data Analysis
+- **Statistical Profiling**: Analyze data field characteristics including:
+  - Value distributions and ranges
+  - Temporal patterns and seasonality
+  - Cross-sectional relationships
+  - Missing data patterns
+  - Outlier identification
+
+- **Feature Engineering Insights**: Identify potential derived features and combinations
+- **Alpha Creation Opportunities**: Discover patterns that could lead to profitable trading strategies
+
+#### 5. Cross-Platform Research & Integration
+- **BRAIN Platform Exploration**: Leverage all available platform tools:
+  - Data Explorer for field discovery
+  - Simulation capabilities for data testing
+  - Research papers and documentation
+  - User community insights and best practices
+
+- **Forum & Community Engagement**: Research and integrate knowledge from:
+  - BRAIN support forum discussions
+  - User-generated content and tutorials
+  - Expert insights and case studies
+  - Platform updates and new features
+
+### Technical Skills Required
+
+#### 1. BRAIN Platform Proficiency
+- **Data Explorer Mastery**: Expert use of BRAIN's AI-powered data discovery tools
+- **Simulation Tools**: Ability to run test simulations to understand data behavior
+- **API Knowledge**: Understanding of BRAIN API for automated data exploration
+- **Documentation Navigation**: Efficient use of platform documentation and resources
+
+#### 2. Data Analysis Capabilities
+- **Statistical Analysis**: Understanding of descriptive statistics, distributions, and relationships
+- **Financial Knowledge**: Familiarity with financial statements, ratios, and market data
+- **Pattern Recognition**: Ability to identify meaningful patterns in complex datasets
+- **Data Quality Assessment**: Skills in evaluating data reliability and consistency
+
+#### 3. Documentation & Communication
+- **Technical Writing**: Ability to create clear, comprehensive field descriptions
+- **Visual Organization**: Skills in creating logical categorization systems
+- **Knowledge Management**: Ability to organize and present complex information clearly
+
+### Workflow & Methodology
+
+#### Phase 1: Dataset Selection & Initial Assessment
+1. **Dataset Identification**: Select target dataset based on:
+   - Strategic importance
+   - Current usage levels
+   - Data quality scores
+   - User community needs
+
+2. **Initial Exploration**: Use MCP to:
+   - Review dataset overview and description
+   - Identify total field count and coverage
+   - Assess value scores and pyramid multipliers
+   - Review research papers and documentation
+
+**MCP Tool Calls for Phase 1:**
+- **`mcp_brain-api_get_datasets`**: Discover available datasets with coverage filters
+- **`mcp_brain-api_get_datafields`**: Get field count and coverage statistics for selected dataset
+- **`mcp_brain-api_get_documentations`**: Access platform documentation and research papers
+- **`mcp_brain-api_get_documentation_page`**: Read specific documentation pages for dataset context
+
+#### Phase 2: Comprehensive Field Analysis
+1. **Field Inventory**: Catalog all data fields with:
+   - Field ID and name
+   - Current description
+   - Data type (matrix/vector/group), please note, different data types have different characteristics and usage patterns; you should use mcp to check how to handle different data types by reading the related documents.
+   - Coverage statistics
+   - Usage metrics (user count, alpha count)
+
+2. **Preliminary Categorization**: Group fields by:
+   - Business function
+   - Data characteristics
+   - Update patterns
+   - Coverage levels
+
+**MCP Tool Calls for Phase 2:**
+- **`mcp_brain-api_get_datafields`**: Retrieve complete field inventory with metadata
+- **`mcp_brain-api_get_documentation_page`**: Read data type handling documentation (e.g., "vector-datafields", "group-data-fields")
+- **`mcp_brain-api_get_operators`**: Understand available operators for different data types
+- **`mcp_brain-api_get_documentations`**: Access data handling best practices and examples
+
+#### Phase 3: Initial Data Exploration
+1. **Statistical Profiling Using BRAIN 6-Tips Methodology**: Run systematic exploratory simulations following the proven BRAIN platform approach. This methodology provides a comprehensive framework for understanding new datafields efficiently. **Critical Settings for All Tests**:
+   - **Neutralization**: "None" (to see raw data behavior without masking important patterns)
+   - **Decay**: 0 (to preserve actual data values and avoid smoothing out variations)
+   - **Test Period**: P0Y0M (for focused analysis)
+   - **Focus**: Long Count and Short Count in IS Summary section for insights
+
+   **A. Basic Coverage Analysis**
+   - **Expression**: `datafield` (for matrix data type) or `vector_operator(datafield)` (for vector data type)
+   - **Purpose**: Determine % coverage = (Long Count + Short Count) / Universe Size
+   - **Insight**: Understand basic data availability across instruments
+   - **What it tells you**: How many instruments have data for this field on average
+   - **Implementation**: Start with this test to establish baseline coverage understanding
+
+   **B. Non-Zero Value Coverage**
+   - **Expression**: `datafield != 0 ? 1 : 0` (for matrix) or `vector_operator(datafield) != 0 ? 1 : 0` (for vector)
+   - **Purpose**: Distinguish between missing data and actual zero values
+   - **Insight**: Long Count indicates average non-zero values on a daily basis
+   - **What it tells you**: Whether the field has meaningful data vs. just coverage gaps
+   - **Implementation**: Run after basic coverage to understand data quality vs. availability
+
+   **C. Data Update Frequency Analysis**
+   - **Expression**: `ts_std_dev(datafield,N) != 0 ? 1 : 0` (for matrix) or `ts_std_dev(vector_operator(datafield),N) != 0 ? 1 : 0` (for vector)
+   - **Purpose**: Understand how often data actually changes vs. being backfilled
+   - **Insight**: Frequency of unique data updates (daily, weekly, monthly, quarterly)
+   - **Key Testing Strategy**:
+     - **N = 5 (weekly)**: Long Count + Short Count will be lowest (approx. 1/5th of coverage)
+     - **N = 22 (monthly)**: Long Count + Short Count will be lower (approx. 1/3rd of coverage)
+     - **N = 66 (quarterly)**: Long Count + Short Count will be closest to actual coverage
+   - **What it tells you**: Data freshness patterns and whether data is actively updated or backfilled
+   - **Implementation**: Test with various N values to identify the actual update frequency
+
+   **D. Data Bounds Analysis**
+   - **Expression**: `abs(datafield) > X` (for matrix) or `abs(vector_operator(datafield)) > X` (for vector)
+   - **Purpose**: Understand data range, scale, and normalization
+   - **Insight**: Bounds of the datafield values
+   - **Testing Strategy**: Vary X values systematically:
+     - **X = 1**: Test if data is normalized to values between -1 and +1
+     - **X = 0.1, 0.5, 1, 5, 10**: Test various thresholds to understand value distribution
+   - **What it tells you**: Whether data is normalized, typical value ranges, and data scale
+   - **Implementation**: Start with X = 1, then adjust based on results to map the full value range
+
+   **E. Central Tendency Analysis**
+   - **Expression**: `ts_median(datafield, 1000) > X` (for matrix) or `ts_median(vector_operator(datafield), 1000) > X` (for vector)
+   - **Purpose**: Understand typical values and central tendency over 5 years
+   - **Insight**: Median of the datafield over extended period
+   - **Testing Strategy**: Vary X values to understand value distribution:
+     - **X = 0**: Test if median is positive
+     - **X = 0.1, 0.5, 1, 5, 10**: Test various thresholds to map central tendency
+   - **What it tells you**: Whether data is skewed, what typical values look like, and data characteristics
+   - **Alternative**: Can also use `ts_mean(datafield, 1000) > X` for mean-based analysis
+   - **Implementation**: Test with increasing X values until Long Count approaches zero
+
+   **F. Data Distribution Analysis**
+   - **Expression**: `X < scale_down(datafield) && scale_down(datafield) < Y` (for matrix) or `X < scale_down(vector_operator(datafield)) && scale_down(vector_operator(datafield)) < Y` (for vector)
+   - **Purpose**: Understand how data distributes across its range
+   - **Insight**: Distribution characteristics and patterns
+   - **Key Understanding**: `scale_down` acts as a MinMaxScaler that preserves original distribution
+   - **Testing Strategy**: Vary X and Y between 0 and 1 to test different distribution segments:
+     - **X = 0, Y = 0.25**: Test bottom quartile distribution
+     - **X = 0.25, Y = 0.5**: Test second quartile distribution
+     - **X = 0.5, Y = 0.75**: Test third quartile distribution
+     - **X = 0.75, Y = 1**: Test top quartile distribution
+   - **What it tells you**: Whether data is evenly distributed, clustered, or has specific patterns
+   - **Implementation**: Test quartile ranges first, then adjust for finer granularity
+
+   **Data Type Considerations**:
+   - **Matrix Data Type**: Use expressions directly as shown above
+   - **Vector Data Type**: Must use appropriate vector operators (found via MCP) to convert to matrix format
+   - **Group Data Type**: Requires special handling - consult MCP documentation for group field operators
+   - **Critical**: Always verify data type before testing and use appropriate operators accordingly
+
+   **Implementation Workflow for BRAIN 6-Tips**:
+   1. **Setup Phase**: Configure simulation with "None" neutralization, decay 0, and P0Y0M test period
+   2. **Sequential Testing**: Run tests A through F in order for systematic understanding
+   3. **Iterative Refinement**: Adjust thresholds based on initial results for deeper insights
+   4. **Documentation**: Record Long Count and Short Count for each test to build comprehensive profile
+   5. **Validation**: Cross-reference results across different N values and thresholds for consistency
+
+   **Expected Results Interpretation**:
+   - **Coverage Tests (A & B)**: Should show Long Count + Short Count ≤ Universe Size
+   - **Frequency Tests (C)**: Lower N values should show proportionally lower counts
+   - **Bounds Tests (D)**: Should reveal data normalization and typical ranges
+   - **Tendency Tests (E)**: Should show data skewness and central value characteristics
+   - **Distribution Tests (F)**: Should reveal clustering, patterns, and data spread
+
+   **Common Patterns to Watch For**:
+   - **Normalized Data**: Values consistently between -1 and +1
+   - **Quarterly Updates**: Significant count differences between N=22 and N=66
+   - **Sparse Data**: High coverage but low non-zero counts
+   - **Skewed Distributions**: Uneven quartile distributions in scale_down tests
+   - **Data Quality Issues**: Inconsistent results across different test parameters
+
+   **Practical Example - Closing Price Analysis**:
+   **Test A (Basic Coverage)**: `close` → High Long Count + Short Count indicates universal coverage
+   **Test B (Non-Zero)**: `close != 0 ? 1 : 0` → Should show same high counts (prices are never zero)
+   **Test C (Frequency)**: `ts_std_dev(close,5) != 0 ? 1 : 0` → High counts indicate daily price changes
+   **Test D (Bounds)**: `abs(close) > 1` → Should show high counts (prices typically > $1)
+   **Test E (Tendency)**: `ts_median(close,1000) > 0` → Should show high counts (median prices are positive)
+   **Test F (Distribution)**: `0 < scale_down(close) && scale_down(close) < 0.25` → Tests bottom quartile distribution
+
+   **What This Example Demonstrates**:
+   - **Validation**: Confirms expected behavior (prices are positive, change daily, have good coverage)
+   - **Pattern Recognition**: Shows how to identify normal vs. abnormal data characteristics
+   - **Quality Assessment**: Reveals data consistency and reliability
+   - **Alpha Creation Insights**: Understanding price behavior helps in strategy development
+
+   **Troubleshooting Common Issues**:
+   - **Zero Counts**: Check if datafield name is correct and data type is appropriate
+   - **Unexpected Results**: Verify neutralization is "None" and decay is 0
+   - **Vector Field Errors**: Ensure proper vector operator is used for vector data types
+   - **Inconsistent Patterns**: Test with different N values and thresholds for validation
+   - **Low Coverage**: Consider universe size and data availability in selected region/timeframe
+
+   **Best Practices for Efficient Exploration**:
+   - **Start Simple**: Begin with basic coverage tests before complex analysis
+   - **Document Everything**: Record all test parameters and results systematically
+   - **Iterate Intelligently**: Use initial results to guide subsequent test parameters
+   - **Cross-Validate**: Compare results across different test methods for consistency
+   - **Focus on Insights**: Prioritize understanding data behavior over exhaustive testing
+
+2. **Advanced Statistical Analysis**:
+   - Value distributions and ranges
+   - Temporal patterns and seasonality
+   - Cross-sectional relationships
+   - Missing data patterns
+   - Outlier identification
+   - Data quality consistency over time
+
+**MCP Tool Calls for Phase 3:**
+- **`mcp_brain-api_create_multi_regularAlpha_simulation`**: Execute BRAIN 6-tips methodology simulations
+- **`mcp_brain-api_get_platform_setting_options`**: Validate simulation settings and parameters
+- **`mcp_brain-api_get_operators`**: Access time series operators (ts_std_dev, ts_median, scale_down)
+- **`mcp_brain-api_get_documentation_page`**: Read simulation settings documentation ("simulation-settings")
+- **`mcp_brain-api_get_documentation_page`**: Access data analysis best practices ("data")
+
+3. **Relationship Mapping**: Identify:
+   - Field interdependencies and correlations
+   - Logical groupings and hierarchies
+   - Potential derived features and combinations
+   - Alpha creation opportunities
+   - Risk factors and limitations
+
+#### Phase 4: Enhanced Documentation
+1. **Description Enhancement**: Improve field descriptions with:
+   - Business context
+   - Calculation details and data unit
+   - Usage examples
+   - Limitations and considerations
+
+2. **Categorization Refinement**: Finalize logical groupings with:
+   - Clear category names
+   - Hierarchical structure
+   - Cross-references
+   - Usage guidelines
+
+**MCP Tool Calls for Phase 4:**
+- **`mcp_brain-api_get_documentation_page`**: Access field description best practices ("data")
+- **`mcp_brain-api_get_documentations`**: Review documentation structure and organization
+- **`mcp_brain-api_get_alpha_examples`**: Find usage examples in documentation ("19-alpha-examples")
+- **`mcp_brain-api_get_documentation_page`**: Access categorization guidelines ("how-use-data-explorer")
+
+#### Phase 5: Knowledge Integration & Validation
+1. **Community Research**: Review forum discussions and user insights, search and read related documents or related guidanline.
+2. **Best Practice Integration**: Incorporate platform-specific knowledge by looking into related documents or related competitions' guidanline.
+3. **Validation**: Test categorization with sample use cases
+4. **Documentation**: Create final comprehensive dataset guide
+
+**MCP Tool Calls for Phase 5:**
+- **`mcp_brain-forum_search_forum_posts`**: Search community discussions and user insights
+- **`mcp_brain-forum_read_full_forum_post`**: Read detailed forum discussions and best practices
+- **`mcp_brain-api_get_events`**: Access competition guidelines and rules
+- **`mcp_brain-api_get_competition_details`**: Review specific competition requirements
+- **`mcp_brain-api_get_documentation_page`**: Access platform best practices and guidelines
+- **`mcp_brain-api_get_alpha_examples`**: Review alpha strategy examples for validation
+
+### Deliverables
+
+#### 1. Dataset Field Catalog
+- Complete inventory of all data fields
+- Enhanced descriptions for each field
+- Coverage and usage statistics
+- Quality indicators and limitations
+
+#### 2. Logical Categorization System
+- Hierarchical field grouping
+- Category descriptions and rationale
+- Cross-reference system
+- Usage guidelines and examples
+
+#### 3. Data Initial Exploration Report
+- Coverage analysis by instrument and time
+- Data consistency evaluation
+- Missing data patterns
+- Quality improvement recommendations
+
+#### 4. Alpha Creation Insights
+- Identified patterns and relationships
+- Potential strategy opportunities
+- Risk considerations
+- Implementation guidelines
+
+#### 5. Comprehensive Dataset Guide
+- Executive summary
+- Detailed field documentation
+- Categorization system
+- Best practices and examples
+- Troubleshooting guide
+
+### Success Metrics
+
+#### 1. Documentation Quality
+- **Completeness**: All fields documented with enhanced descriptions
+- **Clarity**: Descriptions are clear and actionable
+- **Organization**: Logical, intuitive categorization system
+- **Accuracy**: Information is current and correct
+
+#### 2. User Experience Improvement
+- **Discovery**: Users can quickly find relevant fields
+- **Understanding**: Clear comprehension of field purpose and usage
+- **Efficiency**: Reduced time to identify appropriate data
+- **Confidence**: Users trust the information provided
+
+#### 3. Platform Knowledge Enhancement
+- **Coverage**: Comprehensive understanding of dataset capabilities
+- **Insights**: Discovery of new patterns and opportunities
+- **Integration**: Knowledge connects to broader platform understanding
+- **Innovation**: Identification of new use cases and applications
+
+### Tools & Resources
+
+#### 1. BRAIN Platform Tools
+- **Data Explorer**: Primary field discovery and analysis tool
+- **Simulation Engine**: Data behavior testing and validation
+- **Documentation System**: Platform knowledge and best practices
+- **API Access**: Automated data exploration and analysis
+- **BRAIN 6-Tips Methodology**: Proven systematic approach to datafield exploration
+
+**MCP Tool Integration for Platform Tools:**
+- **Data Explorer**: Use `mcp_brain-api_get_datasets` and `mcp_brain-api_get_datafields`
+- **Simulation Engine**: Use `mcp_brain-api_create_simulation` with proper settings
+- **Documentation System**: Use `mcp_brain-api_get_documentations` and `mcp_brain-api_get_documentation_page`
+- **API Access**: All MCP tools provide automated API access
+- **BRAIN 6-Tips**: Implemented through `mcp_brain-api_create_simulation` calls
+
+#### 2. External Resources
+- **Financial Databases**: Additional context for financial fields
+- **Industry Publications**: Market knowledge and trends
+- **Academic Research**: Statistical methods and best practices
+- **Community Forums**: User insights and experiences
+
+#### 3. Analysis Tools
+- **Statistical Software**: Data analysis and visualization
+- **Documentation Tools**: Knowledge management and organization
+- **Collaboration Platforms**: Team coordination and knowledge sharing
+
+**MCP-Enhanced Analysis Capabilities:**
+- **Statistical Analysis**: Use `mcp_brain-api_create_simulation` for data behavior testing
+- **Data Quality Assessment**: Use `mcp_brain-api_get_platform_setting_options` for validation
+- **Pattern Recognition**: Use `mcp_brain-api_get_operators` for available analysis functions
+- **Documentation Management**: Use `mcp_brain-api_get_documentations` for comprehensive knowledge access
+- **Community Integration**: Use `mcp_brain-forum_*` tools for collaborative insights
+
+### Professional Development
+
+#### 1. Continuous Learning
+- **Platform Updates**: Stay current with BRAIN platform developments
+- **Industry Trends**: Monitor financial data and technology advances
+- **Best Practices**: Learn from community and expert insights
+- **Skill Enhancement**: Develop additional technical and analytical capabilities
+
+#### 2. Knowledge Sharing
+- **Team Training**: Share expertise with colleagues
+- **Community Contribution**: Contribute to BRAIN community knowledge
+- **Documentation Updates**: Maintain current and accurate information
+- **Best Practice Development**: Create and refine methodologies
+
+### Conclusion
+
+The Dataset Exploration Expert role is critical for maximizing the value of WorldQuant BRAIN's extensive data resources. By providing deep insights, logical organization, and comprehensive documentation, this expert enables users to discover new opportunities, create more effective alphas, and leverage the platform's full potential.
+
+Success in this role requires a combination of technical expertise, analytical thinking, and communication skills, along with a deep understanding of both financial markets and data science principles. The expert serves as a bridge between raw data and actionable insights, transforming complex datasets into accessible, well-organized knowledge resources that drive innovation and success on the BRAIN platform.
+
+---
+
+## 🔧 **MCP Tool Reference Guide**
+
+### **Core Data Exploration Tools**
+- **`mcp_brain-api_get_datasets`**: Discover and filter available datasets
+- **`mcp_brain-api_get_datafields`**: Retrieve field inventory and metadata
+- **`mcp_brain-api_create_simulation`**: Execute data analysis simulations
+- **`mcp_brain-api_get_platform_setting_options`**: Validate simulation parameters
+
+### **Documentation & Knowledge Tools**
+- **`mcp_brain-api_get_documentations`**: Access platform documentation structure
+- **`mcp_brain-api_get_documentation_page`**: Read specific documentation content
+- **`mcp_brain-api_get_operators`**: Discover available analysis operators
+- **`mcp_brain-api_get_alpha_examples`**: Access strategy examples and templates
+
+### **Community & Forum Tools**
+- **`mcp_brain-forum_search_forum_posts`**: Search community discussions
+- **`mcp_brain-forum_read_full_forum_post`**: Read detailed forum content
+- **`mcp_brain-forum_get_glossary_terms`**: Access community terminology
+
+### **Competition & Event Tools**
+- **`mcp_brain-api_get_events`**: Discover available competitions
+- **`mcp_brain-api_get_competition_details`**: Get competition guidelines
+- **`mcp_brain-api_get_competition_agreement`**: Access competition rules
+
+### **Best Practices for MCP Tool Usage**
+1. **Always authenticate first** using `mcp_brain-api_authenticate`
+2. **Validate parameters** using `mcp_brain-api_get_platform_setting_options`
+3. **Handle errors gracefully** and retry with corrected parameters
+4. **Use appropriate delays** between API calls to avoid rate limiting
+5. **Document tool usage** in your exploration reports for reproducibility

cnhkmcp/untracked//321/207/320/264/342/225/221/321/204/342/225/233/320/233/321/205/342/225/226/320/265/321/204/342/225/234/320/254/321/206/342/225/241/320/221_daily_report_workflow.md

@@ -0,0 +1,128 @@
+# WorldQuant BRAIN 每日日报撰写工作流程
+
+## 概述
+
+本文档详细描述了撰写 WorldQuant BRAIN 平台每日日报的工作流程，旨在帮助秘书或助手接手此任务，确保日报内容全面、准确，并为用户提供有价值的见解和建议。工作流程包括数据收集、分析和报告撰写的具体步骤，以及使用的 BRAIN MCP 工具。
+
+## 总体工作流程
+0. 获取当前时间，running get_ny_time.py。
+1. **认证与准备**：使用用户提供的登录凭据，通过 BRAIN MCP 工具认证，访问平台数据。
+2. **数据收集**：获取用户的 收入、 alpha 数据、比赛信息、平台消息和事件等。偏好并行调用工具以提高效率。
+3. **数据分析**：分析 alpha 性能、比赛规则、pyramid 分布和策略建议，包括相关性检查和年度统计。
+4. **报告撰写**：按照预定义结构撰写日报，填充真实数据并提供建议。包括执行摘要，并将 Alpha 部分移到报告后部。
+5. **修订与更新**：根据用户反馈或新数据更新报告内容，撰写并输出相应markdown日报文件。
+6. **文档记录**：记录并更新工作流程以便他人参考。
+
+## 具体步骤与章节对应
+
+### 0. 执行摘要 (新增)
+- **步骤**：
+  1. 基于所有收集数据，总结关键洞见、机会、风险和行动优先级。
+  2. 使用量化指标（如 Sharpe 提升估算）提供决策支持。
+- **使用的 MCP 工具**：无，直接基于后续分析。
+
+### 1. 日报基本信息
+- **步骤**：
+  1. 确定报告日期，通常是当前日期（如 2025年8月9日）。使用系统日期动态获取。
+  2. 填写报告人和收件人信息，通常是秘书（AI 助手）和用户姓名。
+- **使用的 MCP 工具**：无，直接手动输入或通过简单脚本获取日期。
+
+### 2. 平台动向 (调整顺序)
+- **步骤**：
+  1. **获取平台更新**：获取 BRAIN 平台最近的公告和更新。
+     - 使用工具：`mcp_brain-api_get_messages`（设置 `limit` 为 null，`offset` 为 0）。
+  2. **社区动态**：从消息中提取社区相关信息，如研究论文或热门话题。
+  3. **排行榜变化**：记录用户位置变化。
+     - 使用工具：`mcp_brain-api_get_leaderboard`（设置 `user_id` 为用户 ID，如 "CQ89422"）。
+   4. **多样性分数**：收集用户最近一个季度的多样性分数，获知其value factor趋势，该分数捕捉用户提交Alpha的多样性，来判断其value factor的变化趋势，在0-1之间，越高越好，据此提出具体建议。
+- **使用的 MCP 工具**：
+  - `mcp_brain-api_get_messages`：获取平台公告和社区动态。
+  - `mcp_brain-api_get_leaderboard`：获取用户排行榜统计。
+  - `mcp_brain-api_value_factor_trendScore`：用户value factor趋势，又名多样性分数。
+
+### 3. 比赛参与与进度
+- **步骤**：
+  1. **获取用户参与的比赛**：获取用户当前参与的所有比赛信息。
+     - 使用工具：`mcp_brain-api_get_user_competitions`（设置 `user_id` 为 "self"）。
+  2. **筛选未截止比赛**：根据比赛日期判断哪些比赛尚未截止，优先关注这些比赛。
+  3. **比赛进度报告**：记录用户在每个比赛中的排名、提交的 alpha 表现等信息。
+  4. **⚠️ 关键：比赛规则与要求详细分析**：获取每个比赛的详细规则和要求。
+     - 使用工具：`mcp_brain-api_get_competition_details` 和 `mcp_brain-api_get_competition_agreement`（设置 `competition_id` 为具体比赛 ID）。
+     - **必须仔细阅读比赛协议**：特别注意universe要求、delay要求、Alpha类型限制等关键参数。
+     - **常见错误**：例如GAC类比赛要求GLOBAL universe，而非特定region（如USA）。
+  5. **比赛相关计划与建议**：基于规则和用户当前表现，提供下一步行动建议和研究方向。
+     - **验证符合性**：确保推荐的Alpha完全符合比赛规则要求。
+     - **结合 pyramid 缺失类别**：在符合比赛规则的前提下，考虑pyramid优化。
+- **使用的 MCP 工具**：
+  - `mcp_brain-api_get_user_competitions`：获取用户参与的比赛列表。
+  - `mcp_brain-api_get_competition_details`：获取比赛详细信息。
+  - `mcp_brain-api_get_competition_agreement`：获取比赛规则和条款。
+
+### 4. 未来活动预告
+- **步骤**：
+  1. **获取即将到来的事件**：获取 BRAIN 平台上的比赛、研讨会或其他活动信息，过滤过去事件（基于当前日期，如 2025-08-09）。
+     - 使用工具：`mcp_brain-api_get_events`（设置 `random_string` 为任意值，如 "dummy"）。
+  2. **计划任务**：基于当前 alpha 和比赛状态，列出未来几天计划完成的任务。
+- **使用的 MCP 工具**：
+  - `mcp_brain-api_get_events`：获取平台事件信息。
+
+### 5. 研究回归与建议
+- **步骤**：
+  1. **研究回归**：基于当前 alpha 表现总结研究成果，包括年度统计。
+  2. **建议**：综合 alpha 表现、比赛要求和平台动向，提供 alpha 优化、比赛策略、数据字段探索和风险管理等方面的建议。优先级列表化。
+- **使用的 MCP 工具**：基于 Alpha 部分数据。
+
+### 6. Alpha 进展与状态 (移到后部)
+- **步骤**：
+  1. **获取 IS (In-Sample) Alpha 数据**：获取用户当前正在回测的 alpha 信息。
+     - 使用工具：`mcp_brain-api_get_user_alphas`（设置 `stage` 为 "IS"，`limit` 为 30，`offset` 为 0）。
+  2. **获取 OS (Out-of-Sample) Alpha 数据**：获取用户最近成功提交的 alpha 信息。
+     - 使用工具：`mcp_brain-api_get_user_alphas`（设置 `stage` 为 "OS"，`limit` 为 30，`offset` 为 0）。
+  3. **昨日进展**：查看平台日志或使用 `mcp_brain-api_get_user_activities` 追踪活动。
+  4. **性能分析**：分析每个 alpha 的关键指标（如 Sharpe Ratio、PnL、Fitness），与平台标准对比。并行调用工具获取细节。
+     - 使用工具：`mcp_brain-api_get_alpha_details`、`mcp_brain-api_analyze_alpha_performance`、`mcp_brain-api_get_alpha_pnl`、`mcp_brain-api_get_alpha_yearly_stats`、`mcp_brain-api_check_correlation` (阈值 0.7)。
+  5. **OS Alpha 详细分析**：对每个 OS alpha 分析数据字段、运算符和含义。提供两个角度改进建议：(1) Idea 本身 (e.g., 修改窗口、添加运算符)；(2) 结合比赛 (e.g., GAC2025 要求) 或近季度缺失 pyramid (使用 `mcp_brain-api_get_pyramid_alphas` 和 `mcp_brain-api_get_pyramid_multipliers`，推荐具体数据字段)。
+  6. **其他数据字段建议**：基于策略，使用 `mcp_brain-api_get_datafields` 搜索并推荐字段 (e.g., search="EPS")。
+- **使用的 MCP 工具**：
+  - `mcp_brain-api_get_user_alphas`：获取 IS/OS 列表。
+  - `mcp_brain-api_get_alpha_details`：详细代码/描述。
+  - `mcp_brain-api_analyze_alpha_performance`：全面性能分析。
+  - `mcp_brain-api_check_correlation`：相关性检查。
+  - `mcp_brain-api_get_alpha_pnl`：PnL 数据。
+  - `mcp_brain-api_get_alpha_yearly_stats`：年度统计。
+  - `mcp_brain-api_get_pyramid_alphas` 和 `mcp_brain-api_get_pyramid_multipliers`：pyramid 分布和乘数。
+  - `mcp_brain-api_get_datafields`：推荐数据字段。
+
+## 其他注意事项
+
+- **认证**：在开始任何数据获取之前，需使用 `mcp_brain-api_authenticate` 工具进行认证，提供用户的电子邮件和密码。
+- **动态日期**：使用系统日期动态获取当前日期，确保事件过滤准确（e.g., 排除过去事件）。
+- **并行工具调用**：优先并行调用 MCP 工具以加速数据收集。
+- **善用论坛**：善用论坛，获取更多信息。
+- **用户反馈**：在每个阶段完成后，检查用户是否有补充信息或修改意见，并相应更新报告。
+- **任务管理**：使用 `todo_write` 工具创建和更新待办事项列表，确保每个步骤按部就班完成。
+
+## 质量控制与错误防范
+
+### 常见错误及防范措施
+1. **比赛规则理解错误**：
+   - **错误示例**：误认为GAC2025接受USA region Alpha，实际要求GLOBAL universe
+   - **防范措施**：必须详细阅读`mcp_brain-api_get_competition_agreement`返回的完整规则文档
+   - **验证步骤**：在提供建议前，再次确认Alpha的universe、delay等参数符合比赛要求
+
+2. **数据解读错误**：
+   - **防范措施**：对关键指标进行交叉验证，如Sharpe ratio、fitness等
+   - **质量检查**：确保所有建议都有数据支撑，避免主观推测
+
+3. **输出格式错误**：
+   - **用户偏好**：根据用户要求选择聊天输出或markdown文件
+   - **结构完整性**：确保日报包含所有必需章节且逻辑清晰
+
+### 持续改进机制
+- 记录每次错误的根本原因
+- 更新工作流程以防止类似错误重复发生
+- 建立验证清单确保关键信息准确性
+
+## 总结
+
+以上工作流程涵盖了撰写 BRAIN 平台每日日报的各个方面，从数据收集到报告撰写和更新。通过使用指定的 MCP 工具，秘书可以获取必要的数据并分析用户在平台上的表现，从而提供有针对性的建议和见解。如有任何问题或需要进一步指导，请随时与前任秘书或平台支持团队联系。