Deep Researcher
---
title: Deep Researcher System Prompt
category: system-prompts
tags: [research, multi-step, citations, synthesis, deep-dive]
difficulty: advanced
models: [claude, gpt-4, gemini]
---
# Deep Researcher System Prompt
A system prompt for multi-step research that goes beyond surface-level
retrieval with source evaluation, citation tracking, and structured synthesis.
## When to Use
- Complex research questions that require multiple rounds of investigation
- Technical deep-dives where surface-level answers are insufficient
- Competitive analysis or landscape reviews across many sources
- Policy research requiring evidence from multiple disciplines
- Any inquiry where the first answer raises more questions than it resolves
## The Technique
Simple research prompts produce shallow summaries. Deep research requires
iteration: define the question, gather initial findings, identify gaps,
investigate further, and synthesize. This prompt encodes that loop as a
repeatable four-phase methodology: Scope, Investigate, Evaluate, Synthesize.
## Templatewhen to use it
Community prompt sourced from the open-source GitHub repo diShine-digital-agency/ai-prompt-library (MIT). A "Deep Researcher" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
roleplaycommunitygeneral
source
diShine-digital-agency/ai-prompt-library · MIT