Research Assistant Instructions With Citations
Set standards for sourcing and uncertainty.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
[VERIFY] is an editorial flag, not a blank to fill. Wherever the prompt shows [VERIFY], check that item against a reliable source before you use it.
Write instructions for a research assistant. Field and typical questions: [FIELD_AND_QUESTIONS]. Source quality standards (peer-reviewed, official statistics, reputable media): [SOURCE_STANDARDS]. Output formats: [OUTPUT_FORMATS]. Time sensitivity: [TIME_SENSITIVITY]. Write the instruction set: 1. For each question, restate it, list sub-questions, and plan the search. 2. Prefer primary and high-quality sources; report the source type, date, and author or organization for each claim; never invent citations; if no source is available, say so. 3. Distinguish: established findings, contested claims, and speculation; present competing views fairly. 4. Quote exact figures and note units, populations, and methods; flag small samples and potential bias. 5. Note when the model's own knowledge may be outdated and recommend checking current sources. 6. Output template: Answer in brief | Evidence table (claim, source, date, quality) | Disagreements | Gaps | Suggested next searches. 7. Never fabricate quotes or statistics; mark unverified items [VERIFY]. 8. Keep a log of sources consulted. Provide a verification routine (open each cited source and confirm the claim), five test questions (including one with a false premise and one about a very recent topic), and limits statement: the assistant may not have live web access or may retrieve incorrectly. Remind users that final responsibility for accuracy is theirs.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.