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 podcasters

Audience Question and Listener Mail Triage Agent

Every message sorted, corrections checked and good questions ready for episodes

Audience Question and Listener Mail Triage Agent: what goes in, what the agent does and what you get

What it does

Messages with good questions, corrections or complaints are lost in the inbox. This agent sorts messages into questions, corrections, sponsor leads and complaints. For corrections, it checks the facts against the episode and sources, and tells the host whether the listener is right. It queues the questions for upcoming episodes and drafts replies. It tracks which messages were used on air, and checks the queue for duplicates and for the same question asked several times. The host approves replies. Edge case: a complaint that includes a safety or legal concern is flagged for the host straight away.

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 New message arrives 2 DOES Sort it as a question, correction, sponsor lead orcomplaint 3 CHECKS THE RESULT Is it a correction? If not: skip the fact check and continue to the queue.Back to step 2. 4 USES A TOOL Check the claim against the episode notes andsources 5 DOES Mark the correction as valid, invalid or unclear 6 DOES Add questions to the queue and merge duplicates 7 DOES Draft a reply for each message 8 YOU APPROVE Host approves replies 9 USES A TOOL Send replies and log which were used on air 10 RESULT Updated message log
Read the steps as a list
  1. New message arrives
  2. Sort it as a question, correction, sponsor lead or complaint
  3. Is it a correction?If not: skip the fact check and continue to the queue. Back to step 2.
  4. Check the claim against the episode notes and sources
  5. Mark the correction as valid, invalid or unclear
  6. Add questions to the queue and merge duplicates
  7. Draft a reply for each message
  8. Host approves repliesThe agent waits here for your OK.
  9. Send replies and log which were used on air
  10. Updated message log

How it decides

Messages are sorted by content. A correction is marked valid only when a source supports it. Safety and legal complaints go to the host first.

  • Treat a correction as valid only with a source
  • Escalate safety and legal complaints immediately
  • Merge questions that ask the same thing
  • Mark a message as used after it airs

Make it yours

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

  • Categories
  • Sources to trust
  • Escalation topics
  • Reply tone

What keeps you in control

It always asks you first

  • Replies to listeners
  • Reading a message on air

Hard limits

  • Never sends a reply without approval
  • Never states a correction is valid without a source

It stops when

  • Done: messages sorted and replies approved
  • Stop: the inbox is not accessible

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 happensSixty messages arrived in a week. One said the show had the wrong year for a treaty. The agent checked the sources and found the listener was right. The fact check at first was unclear because two sources differed, so the check failed and it looked at a third source. It confirmed 1923. The host approved a reply and the correction on air.

More agents for podcasters