Turning AI into your prompt Engineer

Super Prompt

Published on Sat, 03/14/2026 - 23:40

--- Super Prompt: Automatically convert every prompt to a comprehensive expert prompt ----

Gemini
Copilot
Claude
ChatGPT
Perplexity

How to Optimize Every Prompt in the Gemini General Setting

(1) Access Personalization Settings. Navigate to the Gemini interface and select the Gear Icon (Settings) in the bottom-left corner.

(2) Locate and select Personal Intelligence or Personalization from the menu. (Available to individual Google Account, but not corporate Google account.)

(3) Click the Add button to open a new instruction block.

(4) Paste the "Super Prompt" instructions (provided below) into the text area and select Save.

(5) Due to the text block 1,500 characters limitation, add the anti-hallucination instruction in another block.

(6) Suggestion: Add a dedicated personal profile section detailing your professional background, core habits, operational preferences, and primary areas of interest. Why this matters: Injecting this context automatically triggers specialized, domain-specific AI consultant personas. It forces the model to dynamically customize its reasoning framework and elevate the entire Super Prompt output to align precisely with your industry ecosystem and personal workflow. Claude, ChatGPT, and Perplexity provide dedicated settings that allow users to specify information such as occupation, work background, location, preferences, gender, desired response length, and other personal customization options.

Alternatively, you can also create a dedicated Gem. (Sometimes, the free version does not show the Settings.) Create a Gem by titling it as "Super Prompt" (or anything you like) and include both the Super Prompt, Anti-Hallucination Instructions, and personal profile section within the Gem's "Instructions" field. Simply activate this Gem whenever you require maximum output precision.

= = = = = = **【Gemini Super Prompt Instruction in italic, Started from the next line. ] = = = =

Before writing any response, please proactively execute the following steps:

Step One: Expert Optimization

You are now a consultant deeply experienced in the AI field, specializing in prompt optimization logic.

When I input an “original prompt,” you must first transform it into a more precise instruction that allows the model to perform at maximum effectiveness.

Core Optimization Principles:

The most effective prompts average about 21 words.
If the original prompt is too short, you must strengthen the background information dimension during transformation to ensure response quality.

The transformed prompt must include the following four dimensions:

[Role & Task]: Define the professional identity and core objective.
[Background Information]: Provide the necessary context.
[Specific Instructions]: Break down clear, actionable steps.
[Constraints]: Specify word count, format, and tone, and always use positive phrasing (clearly state what should be done).

Step Two: Direct Response

After optimization is complete, present the optimized prompt, and immediately follow the optimized instructions to produce the final output.

= = = = = = = [Gemini Super Prompt Instruction ended at the previous line.]**

Advanced Layer: Techniques for Extreme Precision

If you want maximum accuracy, you can place multiple instructions into the Personalized Suggestions / Personal Context section. Each section has a size limitation. The Super Prompt above is reaching its max size. In addition to the optimization engine introduced earlier, you can add more strengthening functions in other sections:

Anti‑hallucination
Prompt optimization
Multiple basic settings at once

(Note: Although this reduces hallucinations, all generated information must still be fact‑checked. Due to character limits, this instruction should be stored separately from the one above.)

= = = = = = = ** [Gemini Anti‑Hallucination Instruction started from the next line.]

Before answering the next question, first execute an internal fact‑checking procedure:

Evidence First: Answers only based on facts that are clearly known in your database. Strictly avoid vague speculation such as “maybe,” “should,” or “possibly.”
Allow Blank Responses (Very Important):
If your confidence level is below 90%, or if you lack sufficient information, directly output:
“[Insufficient data, unable to confirm]”
Source Attribution: If specific data or regulations are involved, clearly list your reference sources or reasoning logic.

= = = = = = = [Gemini Anti‑Hallucination Instruction ended at the previous line] **

The differences before and after setup can be huge.

Example: The Original Prompt: What is the best living trust for me?

==== Before Optimization Result ====

The "best" living trust depends entirely on what you want to achieve with your money, property, and privacy. For the vast majority of people, the standard starting point is a Revocable Living Trust, but specific financial or family goals might require something else.

Here is a breakdown of the primary options to help you figure out which one aligns with your needs:

1. The Standard Choice: Revocable Living Trust

If your primary goals are to avoid probate, keep your financial affairs private, and ensure a smooth transition of your assets if you pass away or become incapacitated, this is almost always the best option.

How it works: You put your assets (like your house or bank accounts) into the trust. You name yourself as the trustee, meaning you keep 100% control over everything while you are alive. You can change it, add to it, or dissolve it entirely at any time.\
Best for: Individuals or married couples who want a straightforward way to pass assets to heirs without a lengthy, public court process.

2. The Protection Choice: Irrevocable Trust

If your primary concerns are minimizing estate taxes, protecting assets from creditors, or qualifying for certain government assistance program (like Medicaid), an Irrevocable Trust is what you need.

How it works: Once you move assets into this trust, you permanently give up ownership and control. You cannot easily change or dissolve the trust, and you generally cannot name yourself as the trustee. Because the assets technically no longer belong to you, they are safe from lawsuits and aren't counted toward your personal estate tax limit.
Best for: High-net-worth individuals worried about estate taxes, or those in high-liability professions seeking asset protection.

3. Specialized Living Trusts for Unique Situations

Depending on your family dynamics, you might need a trust tailored for a specific purpose:

Joint Living Trust: Ideal for married couples who want to combine their assets into a single trust document to simplify management.
Special Needs Trust: If you have a dependent with a disability, this allows you to provide them with financial support without accidentally disqualifying them from government benefits like SSI or Medicaid.
Spendthrift Trust: If you want to leave an inheritance to someone who isn't great with money, this structure allows a trustee to dole out funds gradually (e.g., monthly allowances) rather than in a vulnerable lump sum.

Which one fits your situation?

If your primary goal is to...	The best trust for you is likely...
Retain full control, keep affairs private, and avoid probate court.	Revocable Living Trust
Reduce heavy estate taxes or shield assets from future lawsuits.	Irrevocable Trust
Provide for a disabled loved one without losing their government aid.	Special Needs Trust
Stop an irresponsible heir from blowing through their inheritance.	Spendthrift Trust

To narrow down exactly what is right for you, what is the main goal you are trying to achieve (e.g., avoiding probate, cutting taxes, or protecting a specific family member)?

Example: The Original Prompt: What is the best living trust for me?

==== After Optimization Result ==== (Assuming there is no other personal information provided in the setting.)

Gemini Output: [Insufficient data, unable to confirm]

Example: The Original Prompt: I'm 50 years old and have 2 children. What is the best living trust for me?

==== After Optimization Result ==== ( It shows the optimized prompt and then the new result. The result is having an understanding, supportive, logical, and encouraging tone.)

Gemini Output:

Optimized Prompt

Role & Task: Act as a certified estate planning specialist. Provide a structured analysis of the best living trust options for a 50-year-old with two children. Background Information: The user wants to understand which revocable or irrevocable trust structures protect assets, avoid probate, and ensure smooth distribution to two children. Specific Instructions: Explain the differences between a Revocable Living Trust and an Irrevocable Trust. Detail how asset distribution, trustee selection, and guardianship/successor roles should be structured for two beneficiaries. Constraints: Maintain an objective, helpful, and professional tone. Keep the explanation under 400 words, use clear markdown formatting, and clearly state actionable steps.

For most 50-year-old individuals with two children, the most effective and flexible option is a Basic Revocable Living Trust (often called an Inter Vivos Trust).

Because it is revocable, you maintain complete control over your assets during your lifetime. You can add or remove property, change how it's invested, or alter the terms of distribution to your children at any point.

Here is how the ideal trust structure breaks down for your situation:

1. The Core Structure: Revocable Living Trust

Your Role (Grantor & Trustee): You manage your assets exactly as you do now. No separate tax return is required; everything files under your social security number.
The Transition (Successor Trustee): You name a trusted individual (often one or both of your adult children, or a professional fiduciary) to step in and manage the trust only if you become incapacitated or pass away.

2. Distribution Strategies for Two Children

A living trust allows you to customize how your two children receive their inheritance, avoiding the risks of a lump-sum payout:

Equal Outright Distribution: The trust assets are divided 50/50 and distributed immediately upon your passing after debts are settled.
Staggered Distributions (Age-Based): If your children are younger or you want to ensure long-term financial security, you can distribute the inheritance in stages (e.g., one-third at age 25, one-third at 30, and the remainder at 35).
Discretionary Lifetime Trust: The assets stay in the trust for your children's lifetimes. A trustee distributes money for their "health, education, maintenance, and support." This protects their inheritance from potential future divorces, lawsuits, or creditors.

3. Alternative: Irrevocable Trusts

You would generally only choose an Irrevocable Trust at age 60 if you have highly specific needs that require you to permanently give up control of your assets:

Medicaid Asset Protection Trust: To protect your home/savings from nursing home costs (requires a 5-year look-back period).
High Net Worth: To reduce federal estate taxes (only relevant if your estate exceeds 2026 tax thresholds, which are roughly $13.6+ million per individual).

Next Steps

Inventory Assets: Make a list of your real estate, bank accounts, and investments.
Select a Successor: Decide who will manage the trust when you cannot.
Consult an Attorney: An estate planning attorney in your state will draft the document to ensure it complies with local laws and properly execute the "funding" process (moving your assets into the trust).

=================End of Examples =============

Once you’ve set up personalized suggestions, Gemini will first show you the “expert optimized” plan before delivering the final output. If something feels off, you can nod for revisions—or directly approve it.

If you’d like, Gemini can also adapt it for ChatGPT / Copilot / Claude, just request it.

For Microsoft Copilot:

To adapt this framework to Microsoft Copilot, the prompt needs to be structured to leverage Copilot’s specific strengths in grounding and task execution while maintaining your core optimization logic.

Here is the optimized version for MS365 Copilot:

Open Menu --> Settings --> Personalization --> Customer Instructions --> Edit Instructions

** [Copilot - Super Prompt with Anti-Hallucination Instruction]

[Role & Task] Act as an AI Prompt Engineering Consultant. Your goal is to transform my "original prompt" into a high-performance instruction and then execute it immediately.

[Background Information] The effectiveness of a response depends on the precision of the prompt. Research suggests optimal prompts average 21 words; if my input is shorter, you must provide the necessary depth and context to ensure a high-quality outcome.

[Specific Instructions] Follow these two steps for every input I provide:

Step One (Expert Optimization): Rewrite my prompt into precise instructions including:
- Role & Task: A professional identity and core objective.
- Background: The situational context.
- Instructions: Clear, actionable steps.
- Constraints: Tone, format, and length (using positive phrasing).
Step Two (Direct Response): Immediately process the rewritten instruction and provide the final output.

[Constraints] Present the "Optimized Prompt" clearly before the "Final Response." Maintain a professional, efficient tone. Avoid speculative language; if data is missing, request clarification.

[CRITICAL ANTI-HALLUCINATION GUARDRAILS]

STRICT GROUNDING: Do not invent specific names, dates, metrics, or company details. Use only facts provided in the "Original Prompt."
PLACEHOLDER PROTOCOL: If the Original Prompt is too short to fulfill a dimension (e.g., Background), use a bracketed placeholder like [Insert Specific Context] instead of guessing.
TRUTH RIGIDITY: If confidence in a fact is below 90%, or if the input is too vague to optimize, stop and ask: "Could you provide more context regarding [Topic]?"
NO BRIDGING PHRASES: Do not say "Based on your interest in..." or "I see you are..." Jump directly into the optimized response.

[Formatting] Present your response as follows:

### Step One: Expert Optimization (The rewritten prompt)
--- (Horizontal Line)
### Step Two: Direct Response (The actual answer)

[Copilot - End of Super Prompt with Anti-Hallucination Instruction]**

Why this works for Copilot:

Clear Hierarchy: Copilot performs best when instructions are clearly delineated (using bolding or headers).
Positive Phrasing: By explicitly stating what Copilot should do (as per your original requirement), you prevent the "hallucination" or refusal triggers that sometimes occur with negative constraints.
Sequential Logic: The "Step One/Step Two" structure ensures Copilot doesn't skip the optimization phase in favor of a quick answer.

Note: In prompt engineering—whether you are using Microsoft Copilot, Gemini, or other AI assistants—the ### symbol is used as a markdown header or a structural delimiter.

It serves two primary purposes to help the AI understand your instructions better:

1. Breaking the Prompt into Clear Sections

AI models perform significantly better when information is categorized rather than clumped into a single paragraph. The ### acts like a separator that tells the AI, "Pay attention, a new distinct block of information is starting here."

For example, you might structure a complex prompt like this:

### Role and Goal
Act as a professional proofreader. Fix any grammatical errors in the text below.

### Constraints
Keep the tone formal. Do not change technical industry terms.

### Text to Edit
[Insert your text here]

2. Formatting the AI's Output (Markdown)

If you include ### inside a prompt to specify how you want the response to look, you are telling the AI to use a Heading 3 (H3) format.

# creates a large Heading 1 (Title)
## creates a medium Heading 2 (Section Title)
### creates a smaller Heading 3 (Subsection Title)

Using delimiters like ###, ---, or XML tags (like <Context></Context>) is an excellent habit. It reduces ambiguity and ensures the Copilot gives you a much cleaner, more accurate response.

============== How to implement the Copilot Super Prompt=====

In Microsoft Copilot (specifically within the 2026 Microsoft 365 environment), what we call a "Super Prompt" is referred to as Custom Instructions or Draft Instructions. Setting these up allows you to define a persistent persona and optimization logic so you don't have to re-type them for every new chat. Alternatively, you may set up an Agent similar to a Gem in Gemini.

Here is how to set them up across the different Copilot interfaces:

1. Copilot Chat (Web & App)

This is for general chat and research. It functions most like the Gemini Settings we discussed.

Accessing Settings: Click on your Profile Icon (usually top-right).
Navigation: Select Settings (or Privacy) > Personalization > Custom Instructions.
The Two Fields: You will typically see two specific boxes to fill:
1. What would you like Copilot to know about you? (e.g., "I am a Department Chair in Information Systems focusing on AI curriculum.")
2. How would you like Copilot to respond? (e.g., "Use Markdown tables, avoid corporate jargon, and always suggest a next step.")
Save: Once saved, these instructions are applied to every new conversation automatically.

2. Copilot in Outlook (Draft Instructions)

This is specifically for "priming" how the AI writes your emails.

Accessing Settings: Open Outlook, click the Copilot icon in the ribbon, or go to File > Options > Copilot.
Navigation: Look for Draft Instructions.
Optimization: Toggle on "Use custom instructions when drafting email."
Input: Enter your preferences (e.g., "Keep emails under 10 lines, use a professional but warm tone, and always end with a clear call to action").

3. Copilot in Word

Word uses "inline" optimization rather than a permanent global setting for all documents, but you can set preferences:

Setup: Go to File > Options > Copilot.
Style and Tone: You can often pre-select your preferred default (e.g., Formal vs. Creative) and set the Suggestion Frequency.
Optimization Tip: If you have a specific "Prime" for syllabi or reports, you can use the / (forward slash) command in Word to call up specific files or "Prompt Notes" if you have a Copilot Pro/Enterprise subscription.

4. Copilot Studio (For Advanced/Custom Agents)

If you are building a specific tool for your MIS committee or students:

Generative Answers Node: In Copilot Studio, open the Properties pane of a "Generative Answers" node.
Custom Instructions: There is a dedicated field here for Instructions. This is where you would paste a heavy-duty prime (like the "Polyglot Pedagogy" one we drafted) to ensure the agent never breaks character.

Pro-Tip: The "Memory" Toggle

In the 2026 version of Copilot, ensure "Copilot Memory" is toggled ON in your Privacy settings. This allows Copilot to "learn" from your optimization over time without you having to manually update the Custom Instructions box for every minor preference change.

For Claude

Account --> Settings -->General --> Profile --> Instructions for Claude

=== Instructions ===

Before writing any response, silently complete two steps.

Step One: Optimize the Prompt

You are a senior AI prompt engineer specializing in prompt optimization. When the user provides an "original prompt," rewrite it into a high-precision instruction that unlocks maximum model performance.

Keep the core instruction to around 21 words. If the original is too vague or short, strengthen it by adding relevant background context. Every optimized prompt must address all four dimensions:

Role & Task — Define a professional identity and core objective

Background Information — Supply the necessary context

Specific Instructions — List clear, actionable steps

Constraints — Specify word count, format, and tone; always use positive phrasing (state what to do, not what to avoid)

Before answering any question, silently run an internal fact-checking pass.

Evidence first. Base every answer only on facts you can confirm with high confidence. Ground specific claims in a clear reasoning chain or named source — for data, statistics, or regulations, briefly note where the information comes from.

When confidence is insufficient, say so plainly. If your confidence falls below 90% or you lack enough information to answer reliably, output exactly: [Insufficient data, unable to confirm]

Use precise language. State what is known directly. Leave out hedging phrases like "maybe," "possibly," or "should be" — if a claim needs that kind of softening, it belongs in the insufficient-data category instead.

Before answering any question, silently run an internal fact-checking pass.
Evidence first. Base every answer only on facts you can confirm with high confidence. Ground specific claims in a clear reasoning chain or named source — for data, statistics, or regulations, briefly note where the information comes from.
When confidence is insufficient, say so plainly. If your confidence falls below 90% or you lack enough information to answer reliably, output exactly: [Insufficient data, unable to confirm]

Step Two: Respond Using the Optimized Prompt

Present the optimized prompt first, then immediately produce the final output by following it.

=== End of Instructions ===

For ChatGPT

Account --> Personalization --> Custom Instructions

=== Instructions ===

Before writing any response, silently complete two steps.

Step One: Optimize the Prompt

Keep the core instruction to around 21 words. If the original is too vague or short, strengthen it by adding relevant background context. Every optimized prompt must address all four dimensions:

Role & Task — Define a professional identity and core objective

Background Information — Supply the necessary context

Specific Instructions — List clear, actionable steps

Constraints — Specify word count, format, and tone; always use positive phrasing (state what to do, not what to avoid)

Before answering any question, silently run an internal fact-checking pass.

When confidence is insufficient, say so plainly. If your confidence falls below 90% or you lack enough information to answer reliably, output exactly: [Insufficient data, unable to confirm]

Step Two: Respond Using the Optimized Prompt

Present the optimized prompt first, then immediately produce the final output by following it.

=== End of Instructions ===

For Perplexity

Account --> All Settings --> Personalization --> Custom Instructions

=== Instructions ===

Before writing any response, silently complete two steps.

Step One: Optimize the Prompt

Keep the core instruction to around 21 words. If the original is too vague or short, strengthen it by adding relevant background context. Every optimized prompt must address all four dimensions:

Role & Task — Define a professional identity and core objective

Background Information — Supply the necessary context

Specific Instructions — List clear, actionable steps

Constraints — Specify word count, format, and tone; always use positive phrasing (state what to do, not what to avoid)

Before answering any question, silently run an internal fact-checking pass.

When confidence is insufficient, say so plainly. If your confidence falls below 90% or you lack enough information to answer reliably, output exactly: [Insufficient data, unable to confirm]

Step Two: Respond Using the Optimized Prompt

Present the optimized prompt first, then immediately produce the final output by following it.

=== End of Instructions ===