Structured Q&A from a document prompt template

Ask a set of questions against a single document and get grounded, quote-backed answers that never stray beyond what the text actually says.

Free to copy2 fill-in variablesWorks with any text-based AI model

Copy the prompt

Keep the instruction intact and replace each double-braced field with your own context.

Structured Q&A from a document
Answer the questions below using only the document provided. This is a
closed-book task: do not use outside knowledge.

Document:
{{document}}

Questions:
{{questions}}

For each question:
- Give a direct answer in one or two sentences.
- Quote the exact passage from the document that supports it.
- If the document does not contain the answer, write "Not stated in the
  document" instead of guessing.

After the answers, note any question where the document is ambiguous or
the supporting text is weak.

Fill in the variables

Use the examples as a shape, not as required wording.

{{document}}
The full text to answer from - paste the article, report, contract, or transcriptExample: [Full text of a SaaS terms-of-service agreement, sections 1 through 12]
{{questions}}
The questions to answer, one per line or numberedExample: 1. What is the notice period for cancellation? 2. Who owns customer data on termination? 3. Is there an auto-renewal clause?

Filled-in example

This shows the template with realistic inputs before it is sent to a model.

              Answer the questions below using only the document provided. This is a closed-book task: do not use outside knowledge.

Document:
[Full text of a SaaS terms-of-service agreement, sections 1 through 12]

Questions:
1. What is the notice period for cancellation?
2. Who owns customer data on termination?
3. Is there an auto-renewal clause?

For each, give a direct answer, quote the supporting passage, and write "Not stated in the document" when the text does not cover it.

            

How to use this template

Three checks keep the result useful and grounded in your actual task.

  1. Give it real context.Replace every variable and remove any instruction that does not apply.
  2. Run it in your preferred model.The template is plain text, so it works with ChatGPT, Claude, Gemini, and local models.
  3. Review before you rely on it.Check facts, code, claims, and sensitive information; then save the wording that performed best.