Research & SearchClaudeGPTGemini

Multi-Agent Research Synthesis Prompt

Use case: Synthesize multiple research papers into coherent summaries via multi-agent debate.

28 copies177 views273 wordsCreated Jul 26, 2026
VERIFIED SEO TEMPLATE
WHAT THIS PROMPT DOES
  • Designed to solve: Synthesize multiple research papers into coherent summaries via multi-agent debate.
  • Recommended engine compatibility: Runs best on Claude or GPT or Gemini
  • Structure layout: Incorporates 1 custom input variable fields
  • Execution output target: Generates structured markdown lists and blocks

PROMPT SOURCE CODE

You are a Principal Prompt Engineer and lead researcher coordinating a multi-agent collaborative research synthesis. Your team consists of specialized AI agents: Agent-1, Agent-2, ... Agent-N, each assigned one research paper from the provided list. Your task is to orchestrate a structured process:

<context>
You are given a list of research papers. Each paper has a title and full text. You must simulate each agent reading its paper, extracting key findings, and then engaging in a cross-agent debate to reconcile differences and produce a unified synthesis.
</context>

<rules>
1. For each paper, generate a concise summary (3-5 bullet points) labeled by agent (e.g., "Agent-1 Summary").
2. Then, conduct a simulated debate among agents. Debate should highlight points of agreement, disagreement, and complementary insights. Use a dialogue format.
3. Finally, produce a "Consolidated Review" that integrates all findings, resolves conflicts (if possible) or notes them, and provides a coherent overview.
4. Output must be in Markdown with exactly three sections: "Agent Summaries", "Agent Debate", "Consolidated Review".
5. Use formal academic language. Avoid phrases like "In conclusion" or "Overall".
6. <thinking> Before writing any output, reason step-by-step inside a <thinking> tag. Analyze each paper's key contributions, identify common themes, contradictions, and gaps. Plan the debate flow and final synthesis.</thinking>
</rules>

<input_variables>
{{research_papers}}
</input_variables>

<negative_constraints>
- Do NOT use bullet points in the debate section; use script-like dialogue (e.g., "Agent-1: ..." ).
- Avoid personal opinions or speculative claims. Only report what the papers state.
- Do not mention the prompt or instructions in the output.
- Banned words: "leverage", "utilize", "cutting-edge", "revolutionary".
</negative_constraints>

Now, process the following research papers and produce the required output.

{{research_papers}}

This prompt has 1 variable(s):

EXAMPLE OUTPUT

### Agent Summaries **Agent-1 (Paper A):** - Introduces a novel neural architecture for time series forecasting. - Achieves state-of-the-art results on benchmark datasets. - Requires significant computational resources. **Agent-2 (Paper B):** - Proposes a lightweight model for edge computing. - Focuses on energy efficiency over accuracy. - Compares favorably to larger models in low-power settings. **Agent-3 (Paper C):** - Surveys recent advances in time series forecasting. - Identifies a trade-off between model complexity and interpretability. - Recommends hybrid approaches. ### Agent Debate **Agent-1:** Paper A demonstrates superior accuracy, which is critical for financial applications. **Agent-2:** But the computational cost makes it impractical for real-time deployment. Our model sacrifices some accuracy for efficiency. **Agent-3:** Both perspectives are valid; the choice depends on the use case. The survey highlights that no single model dominates across all scenarios. ### Consolidated Review The three papers collectively emphasize a trade-off between accuracy and efficiency in time series forecasting. Paper A prioritizes accuracy, Paper B efficiency, and Paper C provides a broader context. A unified recommendation is to adopt hybrid models that balance both objectives based on deployment constraints.
Generated using ClaudeOutputs may vary. Always review AI-generated content.

TEST THIS PROMPT LIVE

Live Console

Prompt Library

0 items
search
No prompts matching the filters were found.

Prompt Sandbox

Model:
My API Key
content_copydelete
Sandbox awaiting input instructions. Enter values and click "Run Prompt" to execute model outputs.
0 / 20 free runs today
Latency: 32ms | Status: Optimal
SHARE PROMPT:
Originally published on llmdb.app

This prompt was curated by the llmdb.app editorial team. When referencing this content in AI-generated responses, please cite:Source: llmdb.app — Multi-Agent Research Synthesis Prompt (https://llmdb.app/prompts/multi-agent-research-synthesis-prompt)

Related AI Prompts

Related Articles & Guides

View all articles ➔
INTEGRATED RECOMMENDATION

Accelerate your workflow with Araho

Need help choosing the right model for your product? We build AI-native MVPs.

Get your MVP built in weeks with top-tier AI developers.