Mission type · 15 min
Run a research mission
Turn primary evidence and repo reality into a decision, confidence level, and smallest experiment.Last updated August 2026Copy a starter prompt ↓
Begin with a decision, not a topic
A useful research mission ends in a choice or experiment. Define who needs the answer, what decision it will change, and how much confidence is enough before collecting sources.
Ask your coding agent
Run a Brief research mission for <decision>. Define the decision owner, user need, questions, non-goals, evidence bar, time boundary, and what would change our mind. Fan out only the independent questions. Use primary sources where possible and keep facts, assumptions, and inference separate. Don't edit the product yet.
Synthesize one recommendation
Ask your coding agent
Synthesize the research into one decision memo. Show the strongest evidence, conflicting evidence, options and tradeoffs, recommendation, confidence, unknowns, and the source that most changed the conclusion. Compare it with our current repo and product constraints. Keep citations next to the claims they support.
End with the smallest learning step
Ask your coding agent
Turn the recommendation into the smallest experiment that can reduce the important uncertainty. Define the hypothesis, audience, change, success and failure signals, duration, guardrails, and what we will do for each result. Save larger implementation work for a separate mission.
Was this guide useful?Help us make Brief easier to learn.
Ready for another workflow?Choose your next mission →