Data Preprocessing Pipeline Skill for CSV Quality Checks
Use Claude Code to produce a repeatable CSV cleanup script with explicit duplicate handling, missing-value rules, and an audit summary.
中文Install in Claude Code
Use Python, pandas, and a synthetic CSV with row_id, quantity, and region. Decide the duplicate precedence and imputation rule before running.
- Get the complete `plugins/ai-ml/data-preprocessing-pipeline/skills/preprocessing-data-with-automated-pipelines` directory from the pinned GitHub commit.
- Copy it into the target project's `.claude/skills/`. Check `.claude/skills/preprocessing-data-with-automated-pipelines/SKILL.md`, start Claude Code at the project root, and call `/preprocessing-data-with-automated-pipelines`.
Example task
Create a1 with quantity 2, duplicate a1 with 99, a2 with no quantity, and a3 with 6. Keep the first occurrence and fill the missing value with the median of the retained nonempty values.What to check
The output has three rows and a2 gets quantity 4. The summary records one duplicate removed and one missing value filled.
Original SKILL.md on GitHub · Updated
Frequently asked question
Should deduplication or imputation happen first?
Apply the defined duplicate policy before computing imputation statistics so repeated records do not distort them.