Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for crisis communications managers

Overnight Coverage Triage Agent

Make sure no serious overnight story is missed and the team starts the day with a ranked brief.

MonitoringRisk managementMarketingPR and communicationsManagement
See how it worksGet access
Overnight Coverage Triage Agent: what goes in, what the agent does and what you get

What it does

A story that breaks at 2 a.m. can be everywhere by breakfast, and the comms team learns about it from their boss. Overnight the agent scans news and social media for the brand, its products and named executives. It scores each item by reach, source reputation and risk, and checks whether the source is reliable and whether the story is new or a repeat. It then prepares a short morning brief with the top items, why they matter and a suggested response. If an item passes the high-risk threshold, it wakes the on-call person with a short alert. After the person responds, the agent checks the item again to see whether the story is spreading. Edge case: a satire site story that looks like news. The manager approves any reply.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedNo 1 STARTS WHEN Overnight watch begins 2 USES A TOOL Scan news and social feeds for brand and executivementions 3 DOES Remove duplicates and repeats of earlier stories 4 DOES Score each item by reach and risk 5 USES A TOOL Check the source's reliability and look for a secondsource 6 CHECKS THE RESULT Is the source reliable and is the story confirmedelsewhere? If not: downgrade the score, search again and label theitem unverified. Back to step 4. 7 DOES Send a short alert to the on-call person for itemsabove the threshold 8 USES A TOOL Re-check spread of alerted items every hour 9 DOES Write the morning brief with suggested responses 10 YOU APPROVE Manager approves any reply or statement 11 RESULT Brief delivered and log saved
Read the steps as a list
  1. Overnight watch begins
  2. Scan news and social feeds for brand and executive mentions
  3. Remove duplicates and repeats of earlier stories
  4. Score each item by reach and risk
  5. Check the source's reliability and look for a second source
  6. Is the source reliable and is the story confirmed elsewhere?If not: downgrade the score, search again and label the item unverified. Back to step 4.
  7. Send a short alert to the on-call person for items above the threshold
  8. Re-check spread of alerted items every hour
  9. Write the morning brief with suggested responses
  10. Manager approves any reply or statementThe agent waits here for your OK.
  11. Brief delivered and log saved

How it decides

Each item is scored on reach, source reliability and risk keywords. Above the alert threshold it wakes the on-call person. Below it, the item waits for the brief.

  • Wake the on-call person when risk score is 8 or higher out of 10
  • Score a story as unverified if only one non-major source carries it
  • Treat satire and parody sites as not news
  • Include in the brief any item with reach above 10,000

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Names and topics to watch
  • Alert threshold (default 8 out of 10)
  • On-call contacts
  • Brief time (default 7:00)
  • Trusted source list

What keeps you in control

It always asks you first

  • Manager approves any reply or statement
  • Manager approves changes to the alert threshold

Hard limits

  • Never post or reply on its own
  • Label unverified items clearly

It stops when

  • Done: brief delivered at 7 a.m.
  • Stop: monitoring feed is down for over an hour, alert the on-call person

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensAt 1:40 a.m. a trade blog posts that a supplier recall may affect the brand's flagship product. Score: reach 6,000 and high risk, but only one source. The agent searches again and finds the supplier's own notice, so it raises the score to 8.5 and texts the on-call person. At 3:00 spread has doubled. The brief at 7:00 includes a holding line. The manager approves it.

More agents for crisis communications managers

Communication · PR and communications

Press Inquiry Coordination Agent

Accurate, approved press responses before the journalist's deadline.

7 steps1 check5 jobs
Quality control · PR and communications

Crisis Information Verification Agent

A verified-facts brief that leadership can trust minute by minute.

7 steps1 check6 jobs
Quality control · PR and communications

Media Correction Evidence Agent

Correction requests that are justified and precise.

8 steps2 checks4 jobs
Quality control · Automation

Crisis Holding Statement Readiness Agent

Make every crisis scenario have a current, approved holding statement with reachable approvers.

11 steps2 checks1 job