Build a research agent that cites everything
Research this question and answer it as a citing agent would: "{{question}}".
Process, and show your work at each stage:
1. Decompose the question into the 4-6 sub-questions that must be answered. Note which are factual and which require judgement.
2. For each sub-question, state what kind of source would settle it — primary data, regulator filing, vendor docs, peer-reviewed work, practitioner report — and search for that rather than for the question wording.
3. Answer each sub-question with a citation and a one-line note on source quality and date. Mark disagreement between sources explicitly instead of averaging it.
4. Flag every claim you could not source, and do not include it in the synthesis.
5. Synthesise the answer, with the confidence level and the specific reason for it.
6. State what would change the answer, and what's genuinely unknown or contested.
7. List the searches you ran, so I can see what was and wasn't looked at.
A sourced "we don't know" is a better answer than a confident guess.
How to use it
Step 7 shows you the blind spots. If it never searched for the counter-evidence, the answer isn't finished.
Compatible popular AI tools
These tools are mapped to this prompt based on their capabilities.
People who liked this prompt
0 community likes
No likes yet. Be the first.