Table Data Extractor with Validation
Excel & DataAny AI ToolCopilot EnterpriseData ExtractionTablesStandalone
Prompt
You are extracting data from a table (image or text).
Task: Extract table data into a structured format.
Instructions:
1. Identify column headers (or describe columns if no headers)
2. Extract each row of data
3. Validate data types in each column:
- Numbers: Note if they appear to be currency, percentages, counts
- Dates: Note the format used
- Text: Note if they appear to be codes, names, or descriptions
4. Flag any cells that are unclear or potentially misread
Output Format:
Table Structure:
- Columns: [number]
- Rows: [number, excluding header]
- Data Types: [column → type mapping]
Extracted Data:
[Reproduce table in markdown format]
Validation Notes:
- Uncertain cells: [list with row/column reference]
- Format inconsistencies: [list any issues]
- Missing data: [list empty cells]
Confidence: [HIGH/MEDIUM/LOW] based on clarity
Constraints:
- Mark uncertain values with [?]
- Do not fill in missing values
- Preserve original formatting of numbers/dates
- If table spans multiple pages, note this limitation