Multi-Persona Reasoning
Multi-persona reasoning framework where expert personas (scientist, historian, optimist, devil's advocate) engage in 6-stage brainstorming, critique, and convergence to produce higher-quality analysis than single-perspective prompting.
by sockcymbal
How to install
Open any AI agent and paste this:
Install the Multi-Persona Reasoning persona from https://github.com/sockcymbal/enhanced-llm-reasoning-tree-of-thoughts -- clone the repo, read the setup instructions, ask me for my personal details, replace all template variables, copy the files to the right config locations, and walk me through connecting any integrations it needs.
Your AI handles the rest: downloading files, personalizing the config, and walking you through integrations.
View on GitHubWhat it does
~6-stage reasoning pipeline: brainstorm, critique, refine, converge, validate, synthesize.
~Multiple expert personas (scientist, historian, optimist, devil's advocate) debate each other to stress-test ideas.
~Produces higher-quality analysis than single-perspective prompting by forcing genuine intellectual conflict.
~Works as a system prompt for any LLM. Not tied to a specific agent platform.
~Based on Tree-of-Thoughts research. Structured multi-persona deliberation, not just 'ask three experts.'
Tags
reasoningtree-of-thoughtsmulti-perspectivebrainstorminganalysiscritical-thinking
Works With
Claude
ChatGPT
Claude Code