AI agent for public relations specialists
Media Training Session Preparation Agent
Raise the executive's scores on their own weak spots over a short series of mock rounds.
What it does
Generic training uses the same example questions for every executive, so they leave knowing the theory but not their own weak spots. The agent reads the executive's past interviews, talks and quotes, and the company's current issues. It builds practice questions tailored to that person: their usual deflections, topics they handle poorly and the issues reporters are likely to ask about now. It runs mock rounds in chat, scores each answer for clarity, brevity, accuracy and tone, and notes patterns, such as long answers or a defensive opening. The next round concentrates on the weak spots. After each round the agent compares scores with the previous round. It stops when scores reach the target or after the set number of rounds. Edge case: an executive who does well but leaks an unreleased figure. The specialist reviews the feedback before delivery.
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.
Read the steps as a list
- Training session is booked
- Read the executive's past interviews and quotes
- Identify habits and weak topics
- Build tailored practice questions from habits and current issues
- Run a mock round and ask each question
- Score each answer on clarity, brevity, accuracy and tone
- Do all scores reach 4 out of 5, with no leaked facts?If not: build the next round around the weak scores and repeat. Back to step 4.
- Compare scores across rounds and note the progress
- Draft feedback with examples of better answers
- Specialist reviews the feedback before deliveryThe agent waits here for your OK.
- Feedback report delivered to the executive
How it decides
Questions are chosen from the executive's weakest scores. A round is complete when each answer is scored, and the next one targets scores below 4 out of 5.
- Include at least 3 questions on the executive's weakest topics each round
- Flag any figure not on the approved fact sheet
- Stop after 4 rounds or when all scores reach 4
- Score answers over 60 seconds as too long
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Number of rounds (default 4)
- Score target (default 4 of 5)
- Question types
- Fact sheet used
- Feedback format
What keeps you in control
It always asks you first
- Specialist reviews feedback before delivery
- Specialist approves the final question set
Hard limits
- Never record or share sessions outside the team
- Do not give legal advice
It stops when
- Done: scores meet the target or rounds are used
- Stop: the executive asks to end the session
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.
- 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
