Research · Intermediate

Verified

Automated verification

  • Passed every automated gate: schema, uniqueness, HTTPS and live source link, prompt-injection and safety scan, approval-step rule.
  • Drafting model not recorded.
  • Second-model review: kimi-code/k3.1-pro (no critical or high findings).
  • Verified .
  • Automated checks and AI review only, not human review.
How verification works

Score incoming news items against a fixed question set

Tag each new article on relevance, direction and affected assets at once

Use with BotBento

Opens BotBento with this bot ready to create. You check it, create it in one click, then connect its apps. Nothing runs until you do.

  1. On a scheduleHourly batch of new items arrives
  2. ThenAnswer the fixed question set per item
  3. ResultRanked table plus draft digest

New articles land in a queue through the day. The bot reads each one once and answers the same short, unchanging list of business questions about it, so the reader gets a consistent view rather than a fresh summary every time. Scored items go into a sortable table, with the highest-importance ones surfaced first.

Quick answers

What does it do?

Tag each new article on relevance, direction and affected assets at once. Result: Ranked table plus draft digest.

What does it need?

Tools: Google Sheets, Email, Chat. Intermediate: some setup. The prompt includes an approval step.

How do I run it?

On a schedule: Hourly batch of new items arrives. Copy the prompt below into a bot in BotBento.

Prompt to start with

BotBentoRecommended
You receive batches of incoming news items. For every item, answer the same small fixed set of questions — how important it is, which way it points, and which assets it touches — reading the item once and branching to each question in turn. Include a baseline answer produced with no content attached, so the user can see the model's default lean. Write results to a table sorted by importance. Draft a short digest of the top items and hold it for the user's approval; never send, post, pay or delete anything without their explicit go-ahead.