Extract Structured Data From Messy Text

Pull clean, structured records out of unstructured text into a table or JSON, with unknowns handled honestly.

0 likes 0 dislikes
Sign in to rate this prompt

Prompt

    Extract structured data from the text below.

Source text:
{{text}}

Fields I want for each record (name, and if useful the expected type/format):
{{fields}}

Output format (table or JSON): {{format}}

Rules:
- Produce one row or object per record found, with exactly the requested fields.
- Pull values only from the text — do not infer, normalize, or invent data that isn't stated. If a field is missing for a record, leave it empty / null rather than guessing.
- Preserve values as written unless I asked for a specific format (e.g. dates as YYYY-MM-DD); if you normalize, keep it consistent and say so.
- If the text is ambiguous about a value or a record boundary, flag it instead of quietly picking one.

If a requested field is never present in the text, tell me rather than fabricating it.

Like this prompt?

Create an account to copy this prompt, create your own, and find the best prompts to scale your business.