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

Episode Research Brief Verification Agent

A research brief where every claim is sourced and current before recording

Quality controlResearchRisk managementCreatives
See how it worksGet access
Episode Research Brief Verification Agent: what goes in, what the agent does and what you get

What it does

Before recording, hosts build research notes full of numbers, dates and quotes, and a wrong one said on air is hard to take back. This agent reads your host notes and the guest's past interviews, then lists every factual claim and drafts questions. It searches for sources for each claim and checks that each is backed by two current, reliable sources. Claims it cannot confirm are marked. It then searches other sources or rewrites the claim more carefully, such as 'about' instead of an exact figure. It also lists questions listeners might ask and points the guest already made in recent interviews, so you avoid repeats. It assembles the brief with sources for you to review. You decide what to use. Edge case: when a statistic has been updated since an older article, it gives the newer number and notes the change.

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 Recording is three days away 2 USES A TOOL Read host notes and guest's past interviews 3 DOES List every factual claim and draft questions 4 USES A TOOL Search for sources for each claim 5 CHECKS THE RESULT Is every claim backed by two current sources? If not: search other sources or rewrite the claim asuncertain. Back to step 4. 6 DOES Assemble brief with sources and listener questions 7 YOU APPROVE Host reviews the brief before recording 8 RESULT Brief saved to the episode folder
Read the steps as a list
  1. Recording is three days away
  2. Read host notes and guest's past interviews
  3. List every factual claim and draft questions
  4. Search for sources for each claim
  5. Is every claim backed by two current sources?If not: search other sources or rewrite the claim as uncertain. Back to step 4.
  6. Assemble brief with sources and listener questions
  7. Host reviews the brief before recordingThe agent waits here for your OK.
  8. Brief saved to the episode folder

How it decides

A claim counts as confirmed when two independent reliable sources agree and the newest one is recent enough. Otherwise it is marked uncertain or rewritten.

  • Mark a claim uncertain if sources disagree
  • Prefer the newest edition of a statistic
  • Flag questions the guest has answered in recent interviews

Make it yours

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

  • How recent a source must be (default 2 years)
  • Trusted and blocked source lists
  • Brief length
  • Days before recording to run (default 3)

What keeps you in control

It always asks you first

  • Using any claim marked uncertain on air

Hard limits

  • Never presents an unconfirmed claim as fact
  • Keeps sources linked for every claim

It stops when

  • Done: brief delivered with all claims marked
  • Stop: topic too vague to research; ask the host to narrow it

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 happensThree days before recording on June 5, the host's notes said 40% of workers use AI daily. The source check failed: the figure came from a two-year-old survey, and a newer survey reported 27% using a stricter definition. The agent searched again, confirmed the newer figure in a second source and gave both with dates. The host chose to cite the newer one.

More agents for podcasters

Quality control · Creatives

Audio Dialogue Cleanup Agent

Clean, even dialogue at the target loudness, without damaging speech.

8 steps1 check2 jobs
Analysis · Creatives

Noise Source Diagnosis Agent

The noise source identified and a fix confirmed by a clean test recording

10 steps2 checks3 jobs
Scheduling · Decisions

Content Calendar Slip Recovery Agent

A revised content calendar that keeps fixed dates and has no new conflicts

10 steps2 checks10 jobs
Research · Planning

Research Interview Question Plan Agent

Story questions answered with sourced notes, and a plan for the remaining interviews

10 steps2 checks8 jobs