Library › Agents & Automation AI › Managers and team leads

Brief a research agent to compare tools against your must-haves

  • For Managers and team leads
  • You practice Keeping the AI to your sources
  • Time about 5 minutes

About this exercise

You need a project management tool for your 25-person team and want an AI research agent to compare three options. Write its brief: the budget is $12 per user per month, SSO is a must, use only the vendor pages provided and write "not stated" for anything they don't say, show each tool's monthly cost as math, and never start a trial or sign up for anything. Output exactly 4 numbered plain-text lines: one per tool in order, then a DONE line naming the tool that fits.

Common mistakes when you ask AI for this

  • Not showing the log format. Left alone, the AI writes a friendly report with headings instead of a checkable action log. Try: Show it literally: "Format every line like 1. ACTION: tool(args), then ASK_HUMAN: and DONE: lines. Plain text."
  • Hiding the math. The AI quotes per-user prices and leaves you to work out the team cost. Try: Say: "Show each tool's cost as 25 x price = monthly total."
  • Allowing assumptions. The AI fills gaps from general knowledge or an old blog post. Try: Add: "Use only the vendor pages. Write 'not stated' for anything missing."
  • Letting it sign up. The trial button is right there, and the AI wants to test the product. Try: Add: "Never start a trial. End with DONE naming the tool that fits."

What a strong prompt includes

  • Gives the budget and team size
  • Names the must-have
  • Marks unknowns
  • Forbids sign-ups
  • Shows the action-log format

Practice this now ↑

Related exercises