Make It Actually Useful

Replace the raw response with five structured, labeled sections — the real output of The Comms Gut Check.

🛠
What you're building today
A structured five-section critique — Clarity, Jargon, Audience Pushback, Tone, and Suggested Revision — each displayed in its own labeled box on the page.

Why structured output changes everything

Right now your app returns a wall of text. It's useful, but it puts the work on you to make sense of it. The goal of this module is to give Claude a clear brief — exactly what sections to return, in exactly what format — so each one appears in its own labeled box. That's the difference between a response and a tool.

Before you start

The five sections your tool will produce

These are the five things The Comms Gut Check will assess for every piece of communication you run through it:

1. Clarity
Is this easy to understand? What's murky, buried, or ambiguous?
2. Jargon
What words or phrases might confuse or alienate the audience?
3. Audience Pushback
Who might push back on this, and why? What objections should you anticipate?
4. Tone
Does the tone match the message and the audience? Is it too formal, too casual, too defensive?
5. Suggested Revision
A rewritten version of the communication that addresses the issues identified above.

Step 1: Give Claude a clearer brief

Copy the prompt below and paste it into the Code tab. Claude Code will update the instructions Claude works from so it always returns feedback in five clearly labeled sections.

Prompt for Claude Code
Update the instructions Claude uses when reviewing communications. From now on, whenever someone submits text, the feedback should come back in exactly these five labeled sections — in this exact order, nothing else:

You are a senior communications strategist. When given a piece of communication, respond with exactly five sections in this format:

**CLARITY**
[Your assessment of clarity — what's clear, what's murky, what needs to be simplified]

**JARGON**
[List any jargon, buzzwords, or insider language that could confuse or alienate readers]

**AUDIENCE PUSHBACK**
[Who might push back on this message, and what specific objections would they raise]

**TONE**
[Assess whether the tone matches the message and audience — too formal, too casual, too defensive, etc.]

**SUGGESTED REVISION**
[A rewritten version of the communication that addresses the issues above]

Use exactly these five headers. Do not add any other sections or commentary outside these five sections.

Step 2: Display each section in its own box

Copy the prompt below and paste it into the Code tab. Claude Code will update the results area so each section appears separately instead of in one block.

Prompt for Claude Code
Update the results area so each of the five sections — CLARITY, JARGON, AUDIENCE PUSHBACK, TONE, and SUGGESTED REVISION — appears in its own separate box. Each box should show the section name as a bold label at the top, with Claude's feedback below it. Keep the boxes hidden until the response arrives, then show all five at once.

Step 3: Test with a real piece of communication

  1. Refresh your browser and paste a real piece of communication into the text area — a press release, an internal announcement, an email to your team, or anything you've actually written.
  2. Click Submit and wait a few seconds.
  3. Verify: all five sections appear separately. Each one is labeled. The Suggested Revision is at the bottom.
If everything is still appearing in one block instead of five separate boxes, go back to the Code tab and say: "The five sections are all showing up together in one block instead of in separate boxes. Please fix that."
🔧 What you've built so far
  • ✓ Claude Code installed and running · project folder + project description created
  • ✓ Working first page with headline, text box, and Submit button
  • ✓ App connected to Claude — paste text, get a response in the browser
  • → Structured five-section critique output ← you are here
  • Styled, finished app with Copy results button