Playwright Excel Integration

Overview

Convert Playwright codegen scripts into Excel-driven automations with centralized config and required Playwright CLI validation.

Environment

• Use the playwright conda environment.
• Before running any Python command, run: conda run -n playwright python -c "import sys; print(sys.executable)"
• Do not create or activate any venv or .venv.

Inputs

• Playwright codegen script path
• Excel .xlsx path
• Mapping lines: "hardcoded_value" -> Excel[Sheet][FilterCol==FilterVal][DataCol]
• Optional override: PLAYWRIGHT_TARGET_SUBJECT

Workflow

1. Analyze the Playwright script and the Excel structure (sheets, columns, sample rows).
2. Detect hardcoded .fill() values and confirm that each has a mapping; request clarification for mismatches.
3. Ensure dependencies in the playwright conda env (prefer conda install -n playwright, fall back to conda run -n playwright pip install).
4. Create or update config.yaml using centralized control (paths, patterns, column definitions, constants, tunables, shared texts).
5. Modify the Playwright script:
   • Add a config loader and an Excel loader (polars; see references/excel-loading.md).
   • Replace hardcoded values with data[...].
6. Always run Playwright CLI validation using the Playwright CLI skill before continuing (see references/mcp-validation.md).
7. When refactoring existing pipelines/logic, generate outputs and compare MD5 checksums with reference files (see references/md5-validation.md).
8. Run the updated script with conda run -n playwright python.

References

• references/excel-loading.md
• references/mcp-validation.md
• references/md5-validation.md
• references/examples.md