Two Grok Bots split news monitoring and research reporting
@sensei_prek describes an AI-news radar bot that alerts a second Grok Bot coordinator. The coordinator then researches the item and reports back with a chart; the underlying model claims are not independently verified.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Set up a news-monitoring role that sends relevant leads to a research coordinator. Have the coordinator check primary sources, prepare a concise report and distinguish confirmed facts from rumours or unverified figures.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- @sensei_prek
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The poster reports that a monitoring bot alerted a coordinator, which assembled a research breakdown and chart without him opening tabs. The capture does not independently validate the news, rankings, prices or model-performance claims in that output.
- Human involvement
- The poster configured the two roles and received the report. No source-verification step, approval gate or human validation of the research figures is described.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- The poster set up an AI radar Grok Bot to watch AI news and leak accounts.
- The radar bot detected an item and alerted a second Grok Bot called sensei.
- The coordinator researched the item, gathering benchmark, arena and pricing information into a chart, according to the poster.
Piotr Prekurat, posting as @sensei_prek, describes a two-bot setup for following AI news. One Grok Bot, called AI radar, watches AI news and leak accounts. A second, called sensei, acts as coordinator: it receives alerts, researches the item and reports back to him.
In the reported example, the radar bot spotted a claimed model announcement and pinged the coordinator. Sensei then pulled benchmark information, arena results and pricing and assembled a chart. The poster says he did not open a single tab to get the breakdown. This is a concrete account of a monitoring-to-research handoff rather than two independent chat sessions doing unrelated tasks.
The source’s benchmark numbers, hallucination rates, pricing and launch claims are not independently verified here. They are omitted from the entry’s account of the results. The attached chart was not available for review in the supplied text capture, and the post does not describe a check against primary release documentation.
What the account supports is the division of work: one bot watches for leads and another develops them into a report. It does not establish research accuracy, a reliable news-detection latency or any measured model advantage. The poster’s receipt of a chart is distinct from verification of its contents.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: X · @sensei_prek
Frequently asked questions
How were the two Grok Bot roles divided?
AI radar watched AI news and leak accounts. Sensei received its alert, researched the item and reported back to the poster.
What output did the coordinator reportedly produce?
The poster says it assembled a breakdown and a chart using benchmark information, arena results and pricing.
Does this entry verify the model-performance claims?
No. The figures and launch claims are unverified bot output, not established facts. This entry concerns the reported monitoring and research handoff only.
More details
Remaining steps
- The coordinator reported back to the poster, who says he did not open a tab for the breakdown.
About this account
The reported handoff lets a monitoring bot trigger a separate coordinator to research and summarise an item without a new user request at each stage.
Scout read this post live through the public X API on 2026-10-30 and supplied the verbatim capture with permission to use it instead of reopening X. This entry uses that capture, not an independent retrieval or reproduction. Outcomes are self-reported. The prompt is an editorial suggestion, not a quoted task prompt.
Credit: @sensei_prek
Source published
Added to Agents In Use
Source checked
Tools and topics
Grok Bot
News monitoring · Multi-agent workflow · Research handoff
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.