The tracking plan said "implemented". The dashboard says otherwise. I built a CLI that detects plan↔code drift using only AST.

Описание и применение

CLI‑инструмент plan‑drift обнаруживает расхождения между JSON‑планом аналитики и реальной реализацией в коде Python, выявляя неожиданные события, не реализованные события, несоответствия свойств и динамические случаи без использования LLM.

Задача

Analytics tracking code often drifts from the intended tracking plan, causing silent data quality issues that are hard to detect with IDEs or tests.

Результат

The CLI identifies all mismatches between the tracking plan and code, enabling teams to correct drift before dashboards are corrupted.

Рабочий процесс

• 1. Create a JSON tracking plan file. 2. Run plan-drift --plan tracking-plan.json against the repository. 3. Review the console output for drift categories. 4. Fix code or update the plan based on the report. 5. Re‑run the CLI to confirm resolution. 6. (Optional) Add the CLI to CI to warn on critical drift.

Доказательства

The CLI finds four kinds of drift: UNEXPECTED EVENT, UNIMPLEMENTED EVENT, PROPERTY MISMATCH, DYNAMIC.

Источник

Python 3.11+ is enough. pip install git+... and you're done.

Источник