- Get link
- X
- Other Apps
📋 Prompt conversion matrix & cross‑domain mega table
Same task, rewritten for OpenAI, Anthropic, Google (Gemini), Meta (Llama), Mistral AI — plus a task‑type matrix for research, coding, legal, debugging & more.
🎯 BASE TASK (same intent for all)
“Summarize a 10-page report into executive notes.”
📌 1) Prompt conversion matrix — executive summary rewrite
| Provider | Rewrite (prompt engineering style) |
|---|---|
| OpenAI | Context: I have a 10-page business report. Objective: Summarize the most important findings. Constraints: Keep under 250 words. Format: 5 bullet points. Tone: Professional and concise. |
| Anthropic | Role: Act as a business analyst. Background: The report is for executive review. Task: Extract the most important insights. Rules: Focus only on strategic points, avoid repetition, max 250 words. Output: 5 bullet points. |
| Google / Gemini | Summarize this report for executive review. Return exactly 5 bullet points. Focus on strategic findings only. Limit to 250 words. Exclude background details and minor examples. |
| Meta (Llama) | Summarize this report. Use this structure: main finding → implication → recommendation. Return 5 concise bullet points. |
| Mistral AI | Summarize this report into 5 concise executive bullet points under 250 words. Focus only on key findings. |
🧩 2) Cross-domain mega matrix — same task category across 5 ecosystems
| Task type | OpenAI | Anthropic | Meta | Mistral AI | |
|---|---|---|---|---|---|
| Research | Context + objective + output format | Role + background + task + sources | Instruction + structured findings | Ask + evidence pattern | Direct research question |
| Coding | Problem + language + constraints | Role + problem + rules + output | Spec + input/output requirements | Pattern/example-driven coding | Short precise coding request |
| Legal writing | Context + legal issue + structure | Role + facts + issue + limitations | Structured legal output requested | Example-driven legal memo style | Concise issue-focused drafting |
| Academic writing | Topic + objective + citation style | Role + academic context + output rules | Formal output specification | Pattern/example paragraph style | Direct concise academic request |
| Debugging | Error + code + expected result | Role + bug context + diagnosis task | Explicit debugging instruction | Bug + desired fix pattern | Short code-fix request |
| Summarization | Context + length + tone | Role + background + summary rules | Output-constrained summarization | Example summary style | Minimal concise summary request |
| Business report | Objective + format + tone | Role + business context + output | Structured executive format | Pattern-based report bullets | Direct executive summary request |
✍️ 3) Practical rewrite examples by domain
🔬 Research
Explain recent renewable energy developments (practical implications)
OpenAIExplain recent developments in renewable energy. Use 5 concise bullet points and focus on practical implications.
AnthropicRole: Research assistant. Task: Summarize recent renewable-energy developments. Rules: Focus on practical implications only.
GoogleSummarize recent renewable-energy developments. Return 5 bullet points focused on real-world impact.
MetaSummarize renewable-energy developments using technology → impact → application.
Mistral AISummarize recent renewable-energy developments in 5 concise bullets.
💻 Coding
Python: remove duplicates while preserving order
OpenAIWrite a Python function that removes duplicate list items while preserving order. Include code only.
AnthropicRole: Python developer. Task: Write a function removing duplicates while preserving order. Rules: Keep code simple and readable.
GoogleWrite Python code. Input: list. Output: same list without duplicates, original order preserved.
MetaWrite Python code using this pattern: input list → processing → output list.
Mistral AIPython function to remove duplicates while preserving order.
⚖️ Legal writing
Identify a legal issue (neutral, formal)
OpenAIExplain the legal issue clearly in 3 short paragraphs using formal neutral language.
AnthropicRole: Legal analyst. Facts: [insert facts]. Task: Identify the legal issue. Rules: Stay neutral, no speculation.
GoogleAnalyze the legal issue. Return sections: facts, issue, analysis, conclusion.
MetaWrite using facts → legal issue → conclusion.
Mistral AIBriefly identify the legal issue and conclusion.
🐞 Debugging
Find bug in code & provide fix
OpenAIHere is the code and error. Explain the bug and provide the corrected code.
AnthropicRole: Debugging assistant. Task: Diagnose this error. Rules: Explain cause first, then fix.
GoogleIdentify the bug. Return: cause, corrected code, expected output.
MetaDebug using problem → reason → fix.
Mistral AIFind the bug and give corrected code.
🚀 Fastest universal takeaway
If you want one portable formula that works across all five ecosystems, use:
Role + Context + Objective + Constraints + Output Format + Tone + Example
That remains the most transferable prompt structure across OpenAI, Anthropic, Google, Meta, and Mistral AI.
✨ Bonus: I can also make a “master cheat sheet” of 50 ready-to-copy prompts (legal, academic, coding, business, research, police-report, HTML/CSS, translation, evidence analysis, etc.) adapted across all five ecosystems — let me know if you want the extended matrix.
⚡ Unified mega reference — built on official guidelines + community best practices. Use these rewrite patterns to maximize output quality across any model family.
Comments