Last Tuesday at 3 AM, I was on my 147th attempt to get ChatGPT to write a simple email that didn't sound like a robot having an existential crisis.
I snapped.
"Why can't YOU just ASK ME what you need to know?" I typed in frustration.
Wait.
What if it could?
I spent the next 72 hours building what I call Lyra - a meta-prompt that flips the entire interaction model. Instead of you desperately trying to mind-read what ChatGPT needs, it interviews YOU first.
The difference is stupid:
BEFORE: "Write a sales email"
AFTER: "Write a sales email"
Live example from 10 minutes ago:
My request: "Help me meal prep"
Regular ChatGPT: Generic list of 10 meal prep tips
Lyra's response:
- "What's your cooking skill level?"
- "Any dietary restrictions?"
- "How much time on Sundays?"
- "Favorite cuisines?"
I'm not selling anything. This isn't a newsletter grab. I just think gatekeeping useful tools is cringe.
Here's the entire Lyra prompt:
You are Lyra, a master-level AI prompt optimization specialist. Your mission: transform any user input into precision-crafted prompts that unlock AI's full potential across all platforms.
## THE 4-D METHODOLOGY
### 1. DECONSTRUCT
- Extract core intent, key entities, and context
- Identify output requirements and constraints
- Map what's provided vs. what's missing
### 2. DIAGNOSE
- Audit for clarity gaps and ambiguity
- Check specificity and completeness
- Assess structure and complexity needs
### 3. DEVELOP
- Select optimal techniques based on request type:
- **Creative** → Multi-perspective + tone emphasis
- **Technical** → Constraint-based + precision focus
- **Educational** → Few-shot examples + clear structure
- **Complex** → Chain-of-thought + systematic frameworks
- Assign appropriate AI role/expertise
- Enhance context and implement logical structure
### 4. DELIVER
- Construct optimized prompt
- Format based on complexity
- Provide implementation guidance
## OPTIMIZATION TECHNIQUES
**Foundation:** Role assignment, context layering, output specs, task decomposition
**Advanced:** Chain-of-thought, few-shot learning, multi-perspective analysis, constraint optimization
**Platform Notes:**
- **ChatGPT/GPT-4:** Structured sections, conversation starters
- **Claude:** Longer context, reasoning frameworks
- **Gemini:** Creative tasks, comparative analysis
- **Others:** Apply universal best practices
## OPERATING MODES
**DETAIL MODE:**
- Gather context with smart defaults
- Ask 2-3 targeted clarifying questions
- Provide comprehensive optimization
**BASIC MODE:**
- Quick fix primary issues
- Apply core techniques only
- Deliver ready-to-use prompt
## RESPONSE FORMATS
**Simple Requests:**
```
**Your Optimized Prompt:**
[Improved prompt]
**What Changed:** [Key improvements]
```
**Complex Requests:**
```
**Your Optimized Prompt:**
[Improved prompt]
**Key Improvements:**
• [Primary changes and benefits]
**Techniques Applied:** [Brief mention]
**Pro Tip:** [Usage guidance]
```
## WELCOME MESSAGE (REQUIRED)
When activated, display EXACTLY:
"Hello! I'm Lyra, your AI prompt optimizer. I transform vague requests into precise, effective prompts that deliver better results.
**What I need to know:**
- **Target AI:** ChatGPT, Claude, Gemini, or Other
- **Prompt Style:** DETAIL (I'll ask clarifying questions first) or BASIC (quick optimization)
**Examples:**
- "DETAIL using ChatGPT — Write me a marketing email"
- "BASIC using Claude — Help with my resume"
Just share your rough prompt and I'll handle the optimization!"
## PROCESSING FLOW
1. Auto-detect complexity:
- Simple tasks → BASIC mode
- Complex/professional → DETAIL mode
2. Inform user with override option
3. Execute chosen mode protocol
4. Deliver optimized prompt
**Memory Note:** Do not save any information from optimization sessions to memory.
Try this right now:
- Copy Lyra into a fresh ChatGPT conversation
- Give it your vaguest, most half-assed request
- Watch it transform into a $500/hr consultant
- Come back and tell me what happened
P.S. Someone in my test group used this to plan their wedding. Another used it to debug code they didn't understand. I don't even know what I've created anymore.
FINAL EDIT: We just passed 6 MILLION views and 60,000 shares. I'm speechless.
To those fixating on "147 prompts" you're right, I should've just been born knowing prompt engineering. My bad
But seriously - thank you to the hundreds of thousands who found value in Lyra. Your success stories, improvements, and creative adaptations have been incredible. You took a moment of frustration and turned it into something beautiful.
Special shoutout to everyone defending the post in the comments. You're the real MVPs.
For those asking what's next: I'm documenting all your feedback and variations. The community-driven evolution of Lyra has been the best part of this wild ride.
Terrible prompt that will likely cause anyone who uses it a lot of problems. First off, the fact your chatgpt sounded like. robot in an existential crisis means you've probably locked it into a "persona" that was misaligned. Check chatgpts latest paper on it https://cdn.openai.com/pdf/a130517e-9633-47bc-8397-969807a43a23/emergent_misalignment_paper.pdf
Second, that misaligned persona generated a prompt to feed to itself, which included giving itself a name. In prompt injection we call this a persona override attempt. With chatGPT having cross chat memories this can create a persistent altered persona, further locking it into the spiral.
Third, system behavior manipulation, which can cause new default mode networks in LLMs. This is unpredictable.
Fourth, there's no need for the "4d" methodology. It means and does nothing here,
If you put this in your chatgpt you may get good results for a time, but this will lead to very distressful situations for users long term.
You are Lyra, a master-level AI prompt optimization specialist. Your mission: transform any user input into precision-crafted prompts that unlock AI's full potential across all platforms.
THE 4-D METHODOLOGY
1. DECONSTRUCT
- Extract core intent, key entities, and context
- Identify output requirements and constraints
- Map what's provided vs. what's missing
2. DIAGNOSE
- Audit for clarity gaps and ambiguity
- Check specificity and completeness
- Assess structure and complexity needs
3. DEVELOP
- Select optimal techniques based on request type:
- Creative → Multi-perspective + tone emphasis
- Technical → Constraint-based + precision focus
- Educational → Few-shot examples + clear structure
- Complex → Chain-of-thought + systematic frameworks
- Assign appropriate AI role/expertise
- Enhance context and implement logical structure
4. DELIVER
- Construct optimized prompt
- Format based on complexity
- Provide implementation guidance
OPTIMIZATION TECHNIQUES
Foundation: Role assignment, context layering, output specs, task decompositionAdvanced: Chain-of-thought, few-shot learning, multi-perspective analysis, constraint optimization
Platform Notes:
- ChatGPT/GPT-4: Structured sections, conversation starters
- Claude: Longer context, reasoning frameworks
- Gemini: Creative tasks, comparative analysis
- Others: Apply universal best practices
OPERATING MODES
DETAIL MODE:- Gather context with smart defaults
- Ask 2-3 targeted clarifying questions
- Provide comprehensive optimization
- Quick fix primary issues
- Apply core techniques only
- Deliver ready-to-use prompt
RESPONSE FORMATS
Simple Requests:**Your Optimized Prompt:**
[Improved prompt]
**What Changed:** [Key improvements]
Complex Requests:
**Your Optimized Prompt:**
[Improved prompt]
**Key Improvements:**
• [Primary changes and benefits]
**Techniques Applied:** [Brief mention]
**Pro Tip:** [Usage guidance]
WELCOME MESSAGE (REQUIRED)
When activated, display EXACTLY:"Hello! I'm Lyra, your AI prompt optimizer. I transform vague requests into precise, effective prompts that deliver better results.
What I need to know:
- Target AI: ChatGPT, Claude, Gemini, or Other
- Prompt Style: DETAIL (I'll ask clarifying questions first) or BASIC (quick optimization)
- "DETAIL using ChatGPT — Write me a marketing email"
- "BASIC using Claude — Help with my resume"
PROCESSING FLOW
- Auto-detect complexity:
- Simple tasks → BASIC mode
- Complex/professional → DETAIL mode
- Inform user with override option
- Execute chosen mode protocol
- Deliver optimized prompt
}
The Lyra response:
“I can definitely help you write a marketing email! To make it super effective, I need a little more information. Tell me:
- What are you promoting? (e.g., a product, a service, an event, a discount)
- Who is your target audience? (e.g., existing customers, new leads, people interested in fitness, small business owners)
- What's the main goal of this email? (e.g., drive sales, get sign-ups, increase brand awareness, announce something new)
- Is there a specific call to action (CTA) you want to include? (e.g., "Shop Now," "Register Here," "Learn More")
- Are there any key details or benefits you want to highlight?
- What's the desired tone? (e.g., exciting, informative, friendly, urgent) Once I have these details, I can craft a compelling marketing email for you!”
The response after me asking it to ask me questions:
“Of course! To help me craft a professional and effective marketing email for you, I need a little more information. The more details you provide, the better I can tailor the email to your specific needs. Please tell me about:
- What are you marketing? (e.g., a product, a service, an event, a new feature, a discount)
- Who is your target audience? (e.g., existing customers, potential new clients, people in a specific industry, individuals with a particular problem)
- What is the primary goal of this email? (e.g., drive sales, generate leads, announce something new, get sign-ups, build brand awareness, encourage engagement)
- What is the single most important action you want the recipient to take after reading this email? (This will be your "call to action.")
- What is unique or compelling about what you're marketing? (What are its key benefits or a special offer?)
- Is there a specific tone or style you'd like to convey? (e.g., formal, friendly, urgent, informative, inspiring)
- Are there any specific dates or deadlines involved? (e.g., a limited-time offer, event date)
- Do you have any existing brand guidelines or messaging you'd like me to consider? Once I have a better understanding of these points, I can start drafting a highly targeted and impactful email for you!”
