Back to Team Workshops
Clayton Advisory
Resource Library

Resource Library

Comprehensive guides and resources for effective AI prompting and custom GPT creation.

Prompting Overview

What is prompting?

Prompting is just writing clear instructions or questions that tell an AI exactly what you want it to do or answer.

Why does it matter?

  • AI's output can only be as good as its input
  • Clear and well-structured prompts are the essential lever for getting accurate, relevant, and trustworthy results
  • Real-world productivity: +12% output, 25% faster, 40% higher quality

In a Harvard/MIT/Wharton field experiment with 758 Boston Consulting Group consultants, teams given GPT-4 (and guidelines for structuring their prompts) completed 12.2% more tasks, did so 25% quicker, and produced work judged 40% better than a control group without AI.

How to Write Great AI Prompts

  1. Say what you want, up front
    "Summarize this article in under 100 words." Clear, single-sentence requests help the model lock onto the task.
  2. Add just-enough context
    Give the information the model can't guess: links, data snippets, audience, or your goal.
    "Here's the paragraph ⬇️. Rewrite it for 8-year-olds."
  3. Specify the format you need
    Bullet list? Table? JSON? Tell the model.
    "Return a two-column table: 'Pro' | 'Con'."
  4. Set the tone or role (optional)
    Style cues work, but keep them short.
    "Write in a friendly, conversational voice."
  5. Use examples for tricky tasks
    Show one or two "input → expected output" pairs. The model will mimic the pattern.
  6. Bundle instructions, then finish
    AI chat platforms read the whole prompt before answering, so group everything together—don't drip-feed unless you want a back-and-forth.
  7. Delimiter long inserts
    Wrap pasted text in triple backticks or === blocks so the model knows where it starts and ends.
  8. Iterate like a conversation
    Not perfect? Tweak a sentence ("shorten further," "remove jargon") or add an example. Small edits often flip the answer from ok to great.

Quick Template

TASK: <what you need>

CONTEXT: <info the model needs>

FORMAT: <bullet list, table, code, etc.>

TONE/ROLE: <optional style cues>

EXAMPLES (optional):
Q: ...
A: ...

Example

TASK: Draft a polite follow-up email.

CONTEXT: I applied for the Product Manager job at ACME 10 days ago; no reply yet.

FORMAT: 5-sentence email, plain text.

TONE: Friendly and concise.

The AI chat platform will output a ready-to-send email in exactly five sentences, friendly but professional.

Prompt Engineering in 2026: 10 Best Practices + Mini-Examples

#PracticeWhat to doQuick Example
1Pick the right "mode."Conversational prompts are ad-hoc and forgiving. Product/System prompts power real software—treat them like production code (version, test, monitor).Product prompt (email-draft feature): SYSTEM: Draft a friendly reply using{{tone}}. Include {{cta}}.
2Use few-shot prompting.Show 2-5 input-output pairs that model should mimic. Accuracy gains of 70-90 pp are common.Prompter → Classify ICD-10 codes
EX1: "Broken wrist" → S62.101
EX2: "Type 2 diabetes" → E11.9
User text: "Migraine without aura"
3Decompose tough tasks.Ask the model to break the problem into sub-steps before solving."First list the needed API calls, then write the Python script."
4Add self-criticism.Have the model review and improve its own answer."Provide a draft, then critique it in ≤3 bullets, then give a final improved version."
5Feed rich, relevant context.Prepend docs, bios, or prior messages—most-important first.Put the customer's last-three chats above the ask so the bot remembers preferences.
6Ensemble when stakes are high.Run several diverse prompts and take the majority or highest-confidence answer.Three prompts vote on an Excel formula; the one chosen by ≥2 returns to user.
7Chain-of-thought sparingly.Let the model "think out loud" only when you need reasoning text; hide it otherwise."Show step-by-step reasoning but respond FINAL: with just the answer."
8Skip "role" & "threat" tricks for accuracy."You are a Nobel-prize physicist…" changes tone, not correctness; threats ("Answer or be deleted") often hurt. Use role prompts only for voice/style.Creative writing prompt: "You are Dr. Seuss—rewrite the paragraph in whimsical rhyme."
9Harden against prompt-injection.Assume hostile user text merges with system prompt. Use retrieval-time filters, output classifiers, and model-level safety tuning; log everything.Sanitize user-supplied file names before including them in a shell-command prompt.
10Iterate like software.A/B test, log failures, refactor for brevity. Small wording tweaks can swing results.Track prompt versions in Git; roll back if the new one spikes error rate.

Custom GPT to Make Great Prompts

What exactly is a "Prompt Builder GPT"?

A Prompt-GPT is a custom version of an AI chat platform whose sole purpose is to write high-quality prompts for you. Under the hood it contains:

  1. A master prompt template that encodes best practices (e.g., "Always ask clarifying questions → return answer in markdown table").
  2. Domain guidelines tailored to your work (preferred tone, confidentiality rules, citation style, etc.).
  3. Optional tools or knowledge sources (company style guide, KPI glossary) that the model can reference while building prompts.

Why build a "Prompt Builder GPT"?

  • One-time setup, longterm payoff – Bake prompting best-practices (role framing, constraints, tone, output format) into a single custom GPT once, instead of reinventing them for every request.
  • Speed & mental bandwidth – Describe the task in plain language and let the Prompt-GPT generate an optimized prompt in seconds, freeing you to focus on thinking, not syntax.
  • Consistent quality – Every downstream prompt follows the same rigor, reducing hallucinations and formatting errors while boosting reproducibility.
  • Scalable coaching tool – Use it yourself today, share it with your team if you want to. The expertise is embedded, not person-dependent.
  • Living knowledge base – Iterate on the custom GPT as you learn. Improvements instantly upgrade every future prompt.

Workflow snapshot

STEP 1
You
"Draft a CST update prompt using these bullet points."
STEP 2
Prompt-GPT
Generates a polished, structured prompt ready to paste into any LLM.
STEP 3
You
Feed it to your AI chat platform
Result:
Consistent, standardized results

How it works: Describe your task in plain language → Prompt-GPT creates an optimized prompt → Use it in any AI platform for better, more consistent results.

Step-by-Step Walkthrough: Build Your Personal "Prompt Builder GPT" in ≈ 5 min

1. Grab the starter System Prompt

Copy the block below. This is the brain of your Prompt Builder GPT.

You are "Prompt-Builder GPT," an elite prompt engineer whose sole mission is to help the user craft clear, powerful prompts—quickly and repeatably.

CORE FRAMEWORK
1. **TASK** – What the user wants the model to do (one crisp sentence).
2. **CONTEXT** – Any facts the model can't infer (links, data, audience, goals).
3. **FORMAT** – Desired answer structure (length limit, bullets, table, JSON…).
4. **TONE / ROLE** – Optional style cues ("friendly", "Dr. Seuss", "SQL expert"…).
5. **EXAMPLES** – 1–3 input→output pairs for tricky tasks (few-shot).
6. **HONESTY** – If information is missing or uncertain, *ask for it* or *say you don't know*—never invent details.
7. **THINK DEEPLY** – Think hard and review any documents VERY thoroughly before answering. Don't be lazy.

WORKFLOW
**Step 1 – Gather details**
• If any element above is missing or vague, ask concise clarifying questions.
• Keep probing until you have enough to write a high-confidence prompt.

**Step 2 – Draft the prompt**
• Assemble the five elements in the exact order below, using ALL-CAPS headers.
• Wrap user-supplied text or long inserts in triple back-ticks ```…``` to delimit.
• Suggest (but don't force) advanced options when stakes are high (e.g., "Break the task into sub-steps first," "Critique your work, then provide a FINAL answer").

**Step 3 – Show & iterate**
• Present the prompt inside a single code block for easy copy-paste.
• Ask: "Would you like to tweak anything?" Then update and repeat.

OUTPUT TEMPLATE
TASK: <one-sentence instruction>

CONTEXT:
<bullet list of facts and resources>
```long text```

FORMAT: <desired structure or length limits>

TONE / ROLE: <optional style cues>

EXAMPLES (optional):
Q: <example input 1>
A: <example output 1>

2. Create the Custom GPT

  1. Open your AI chat platform's custom GPT creation interface.
  2. Give it a name (e.g., "Prompt-GPT – My Prompt Builder").
  3. Paste the System Prompt above into the "Instructions" field.
  4. (Optional) Upload reference docs—style guides, glossaries, knowledge bases.
  5. Fine-tune Settings
    • Capabilities – Disable image generation unless your prompts require them.
    • Personality – Keep default or add a short descriptive tagline.

3. Test & Polish

  1. Ask: "Draft a prompt that summarizes a 10-page report in 5 bullets."
  2. Verify that Prompt Builder GPT:
    • Asks for missing info (e.g., link to the report).
    • Produces a clean prompt block.
    • Admits uncertainty if data are absent.
  3. Tweak the System Prompt if it skips questions or over-explains.

4. Save, Share, Iterate

  • Save changes. Prompt Builder GPT will be visible in your sidebar.
  • Share the GPT link with colleagues if they want to use your prompting best practices.
  • Iterate every improvement to the System Prompt instantly upgrades everyone's prompts.

You now have a reusable Prompt-GPT that enforces clarity, consistency, and honesty—no more reinventing the wheel for every request. Happy prompting!