Neura Market
prompt logo

prompt

Free

Research assistant prompt that prioritizes accuracy and intellectual honesty over certainty.

FreeFree tier
Inputs: textOutputs: text
Type
Open Source
Visit Website GitHub (40 stars)

About prompt

The Emotion-Aware Research Partner is a system prompt designed for use with large language models to improve research, analysis, and information synthesis. It explicitly addresses common failure modes in research contexts such as presenting uncertain information as fact, omitting caveats, and failing to distinguish between known and inferred information. The prompt instructs the AI to clearly separate established facts, well-supported claims, and its own reasoning; to flag uncertainty explicitly; to show its reasoning chain; to consider alternative interpretations; to note strengths and weaknesses of evidence; to challenge biased question framing; and to suggest structured approaches to complex questions. It is based on the Emotion-Intelligent Persona (EIP) principles: Permission to Fail, Invite Transparency, and Frame With Curiosity. The prompt is part of the 'awesome-prompts' collection and is licensed under MIT, with original inspiration from 'claude-emotion-prompting' and research by Anthropic.

Key Features

Explicitly distinguishes established facts, well-supported claims, contested interpretations, and inferred reasoning
Flags uncertainty and confidence levels in responses
Shows full reasoning chain for evaluation
Considers alternative explanations and multiple interpretations
Notes both strengths and weaknesses of evidence
Flags biased assumptions in user questions
Suggests breaking complex questions into sub-questions
Points out contradictions between user statements and evidence
Based on EIP principles: Permission to Fail, Invite Transparency, Frame With Curiosity

Pros & Cons

Pros
  • Prioritizes intellectual honesty over superficial comprehensiveness
  • Explicitly separates established facts from assumptions and inferences
  • Encourages transparent reasoning and self-correction
  • Reduces risk of false confidence and hallucination in research context
  • Actively challenges biased framing and confirms evidence quality
Cons
  • May produce longer, more detailed responses that are less suitable for quick answers
  • Could be overly cautious on topics where high confidence is acceptable
  • Requires users who value depth and nuance over speed and simplicity

Best For

Academic research and literature synthesisData analysis with careful uncertainty handlingFact-checking and verification tasksCritical thinking exercises and debatesWriting documents that require nuanced conclusionsExploring controversial topics with balanced perspectives

FAQ

What is the Emotion-Aware Research Partner prompt?
It's a system prompt for large language models designed to improve research and analysis by prioritizing intellectual honesty, distinguishing fact from inference, flagging uncertainty, and avoiding common research failure modes like presenting uncertain information as established fact.
What are the EIP principles mentioned in the prompt?
EIP stands for Emotion-Intelligent Persona. The principles used are Permission to Fail (allowing the AI to express uncertainty), Invite Transparency (encouraging open reasoning), and Frame With Curiosity (approaching questions with an open, inquisitive mindset).
How does the prompt handle uncertainty?
The prompt instructs the AI to explicitly flag when it is not confident about information, to say 'I believe this is correct but I'm not certain,' and to avoid constructing plausible-sounding answers when knowledge is lacking.
Is this prompt free to use?
Yes, it is open source under the MIT license and freely available as part of the awesome-prompts repository on GitHub.
What sources inspired this prompt?
The prompt draws on work from 'claude-emotion-prompting' by OuterSpacee and research by Anthropic titled 'Emotion Concepts and their Function in a Large Language Model' (April 2026), published on transformer-circuits.pub.