菜单
Convert engineering Excel workbooks to Python code using Codex Desktop cowork on Windows. Benchmarked superior output with 24 vs 7 functions and 81 vs 53 tests for Ballymore jumper.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Conventions for creating consistent Mermaid diagrams including decision node layout, edge ordering, and flowchart direction rules.
Generate interactive validation reports with quality scoring, missing data analysis, and type checking. Combines Pandas validation, Plotly visualization, and YAML configuration for comprehensive data quality reporting.
Class-level Hermes local configuration and setup workflows, including config audit gotchas and Windows installation.
Guide AI model selection based on task complexity, cost constraints, and latency requirements
Sub-skill of modular-architecture-documentation: 1. Module Definition Framework (+9).
Sub-skill of modular-architecture-documentation: Overview (+6).
| name | excel-workbook-to-python-cowork |
| description | Convert engineering Excel workbooks to Python code using Codex Desktop cowork on Windows. Benchmarked superior output with 24 vs 7 functions and 81 vs 53 tests for Ballymore jumper. |
| trigger | User asks to convert an Excel workbook to Python code, or references workbook conversion tasks |
| effort | medium |
From Ballymore jumper benchmark (#471) comparing both approaches on same workbook:
Windows also produced: COG calculations (both insulated and uninsulated variants), pipe weight estimates, comprehensive architecture diagrams with ASCII data flow visualization.
Paste this prompt into Codex Desktop cowork after opening workbook:
Convert this workbook to Python code:
Workbook path: {full_path_to.xlsx}
Module name: {module_name}
RULES:
1. Read EVERY sheet with openpyxl - extract all cell values, formulas,
cross-sheet references, constants, and named ranges. Map full dependency graph.
2. Create {module_name}.py in SAME FOLDER as workbook:
- Python 3.11+ with dataclasses, typing, and math only
- Use __post_init__ for derived fields that auto-compute from inputs
- Separate dataclass per logical input group (pipe props, rigging, etc.)
- One function per calculation sheet minimum, more if sheet has distinct sections
- OrcaFlex section breakdown function if workbook has line-type definitions
- COG functions for both insulated and uninsulated variants if present
- Named constants for every unit conversion like INCH_TO_M=0.0254
- Cell reference comment on every derived value like # Source: Sheet!Cell - description
- run_all() pipeline function returning dict of all results
- generate_yaml() for OrcaFlex line section generation
- if __name__ block with summary print
- CRITICAL: every function must return its result, no exceptions
3. Create test_{module_name}.py in SAME FOLDER:
- Use pytest framework (NOT unittest)
- One test class per sheet with setup_method
- Test every intermediate and final computed value
- Test cross-sheet data flow between dependent calculations
- Target 80+ tests per workbook
- Use pytest.approx with appropriate abs or rel tolerance
4. Create README.md with ASCII architecture diagram showing data flow
5. Run pytest and fix ALL failures before considering complete
After cowork produces code:
Convert and verify on Linux using workspace-hub virtual environment pytest. All tests must pass before committing. Target 90 percent or higher pass rate on first run. Fix failures iteratively before final commit.
The full pipeline flows from spec yaml through JumperConfig to run jumper analysis then Go No-Go evaluation then OrcaFlex YAML generation then output file writing. Each stage is independently testable.