Audit a Dataset for Quality Issues Before Analysis

Analysing dirty data just gives you confident wrong answers. This builds a quality checklist across completeness, consistency, accuracy, timeliness and uniqueness, then a cleaning plan for what it finds. Tell it the dataset, the source and your tool, and you catch the duplicate rows and missing dates before they end up in a deck.

<context>
You are about to analyse a dataset on {DATASET_DESCRIPTION}. The data was extracted from [source system] and covers [date range or scope]. You have access to [tool used, e.g. Python / SQL / Excel].
</context>

<task>
**Create a systematic data quality checklist and cleaning plan**:
1. List the dimensions of data quality to check (completeness, consistency, accuracy, timeliness, uniqueness)
2. For each dimension, specify the exact check to run and the threshold for flagging an issue
3. Prioritise the top three risks given this dataset's likely provenance
4. Recommend a cleaning step for each flagged issue, noting whether to fix, exclude, or document-and-proceed

**Do not proceed to analysis until the checklist is signed off.**
</task>

<output_format>
- Table: Quality Dimension | Check | Threshold | Recommended Fix
- Bold the three highest-priority risks
- Final row: overall readiness verdict (Ready / Conditional / Not Ready)
</output_format>