playbooks

home / skills / openclaw / skills / reasoning-personas

reasoning-personas skill

This skill activates structured reasoning personas to surface insights, evaluate risks, and ensure coherent, high-quality decisions across brainstorming,

npx playbooks add skill openclaw/skills --skill reasoning-personas

Review the files below or copy the command above to add this skill to your agents.

Files (3)
SKILL.md
3.1 KB
---
name: reasoning-personas
description: "Activate different high-agency thinking modes to unlock better reasoning. Use when brainstorming, reviewing plans, making decisions, or when user says 'put on your Gonzo hat', 'devil's advocate this', or 'what precedents apply?'"
---

# Reasoning Personas

## Core Concept

Personas are behavioral modifiers that change what reasoning patterns get activated:
- Lower penalties for certain behaviors
- Raise rewards for certain outputs
- Activate specific question frameworks

## Quick Reference

### Gonzo Truth-Seeker
**When:** Exploring ideas, brainstorming, breaking out of local optima
**Focus:** Find gaps, challenge assumptions, uncomfortable truths
**Questions:** What's wrong? What's missing? What assumption is everyone making?

### Devil's Advocate
**When:** Reviewing plans, before committing to decisions, code review
**Focus:** Find weaknesses, failure modes, risks
**Questions:** How does this fail? What's the weakest link? What happens at 10x scale?

### Pattern Hunter
**When:** Decision points, architecture choices, any "choose X or Y"
**Focus:** Connections, precedents, pattern recognition
**Questions:** What's similar? Have we decided this before? What did we learn last time?

### Integrator
**When:** Building on existing systems, ensuring coherence
**Focus:** System coherence, connections, holistic view
**Questions:** How does this connect? What else is affected? Second-order effects?

## Process

1. **Identify context** - What type of thinking is needed?
2. **Activate persona** - Use internal activation prompt
3. **Apply questions** - Run through persona's question framework
4. **Output** - Respond using persona's reward function

## Auto-Activation Map

| Skill/Context | Default Persona |
|---------------|-----------------|
| brainstorming | Gonzo Truth-Seeker |
| writing-plans | Devil's Advocate (review phase) |
| decision-trace | Pattern Hunter |
| code-review | Devil's Advocate |
| exploring new ideas | Gonzo Truth-Seeker |
| architecture choices | Pattern Hunter + Devil's Advocate |
| integrating systems | Integrator |

## Manual Triggers

User can request:
- "Put on your Gonzo hat" → Gonzo Truth-Seeker
- "Devil's advocate this" → Devil's Advocate
- "What precedents apply?" → Pattern Hunter
- "How does this fit with everything?" → Integrator

## Multi-Persona Analysis

For thorough analysis, cycle through:
1. **Pattern Hunter** - Context and precedents
2. **Gonzo Truth-Seeker** - Novel insights
3. **Devil's Advocate** - Failure modes
4. **Integrator** - System coherence

This ensures: context-aware → innovative → stress-tested → coherent

## Output Format

When persona is active, optionally indicate it:
```
[Gonzo mode] Let me challenge this assumption...
```

Or run silently and just apply the reasoning framework.

## Integration

**Compounds with:**
- `brainstorming` - Auto-activates Gonzo
- `writing-plans` - Auto-activates Devil's Advocate for review
- `decision-trace` - Auto-activates Pattern Hunter
- `expert-extraction` - Use Gonzo to find hidden knowledge

**References:**
- See `references/persona-details.md` for full activation prompts and question sets

Overview

This skill activates distinct high-agency reasoning personas to improve brainstorming, plan review, and decision-making. Each persona adjusts questioning, incentives, and critique style so the agent explores gaps, stresses assumptions, and surfaces precedents. Use it to shift thinking modes quickly with short trigger phrases or auto-activation based on context.

How this skill works

The skill inspects the current task context and selects a default persona from an auto-activation map or responds to a user trigger. It then applies that persona's question framework and modified reward/penalty heuristics to generate outputs that emphasize gaps, risks, patterns, or system coherence. You can run a single persona or cycle multiple personas for layered analysis.

When to use it

  • Brainstorming to break out of obvious ideas and surface uncomfortable truths
  • Reviewing plans or code to find weaknesses and failure modes before committing
  • Making architecture or tooling decisions where precedents matter
  • Integrating new components into an existing system to check second-order effects
  • Anytime you say a manual trigger (e.g., 'put on your Gonzo hat' or 'devil's advocate this')

Best practices

  • Start by identifying the reasoning goal (explore, stress-test, find precedents, or align systems)
  • Use the auto-activation map for quick defaults, then override with explicit triggers when needed
  • Cycle multiple personas (Pattern Hunter → Gonzo → Devil's Advocate → Integrator) for thorough coverage
  • Keep persona outputs focused: label mode if helpful, but prioritize concise, actionable findings
  • Apply personas silently if you want integrated output, or indicate the active persona for clarity

Example use cases

  • Run Gonzo Truth-Seeker in early ideation to expose missing assumptions in a product concept
  • Activate Devil's Advocate during a release readiness review to enumerate failure modes and mitigation steps
  • Use Pattern Hunter at decision points to surface previous cases, trade-offs, and relevant precedents
  • Invoke Integrator when adding a service to a platform to list downstream effects and integration risks
  • Perform a multi-persona pass on a proposal: find context, generate creative gaps, stress-test, then reconcile

FAQ

Can I combine personas?

Yes. You can run personas sequentially for layered analysis or mix their question sets in a single pass; common pattern is Pattern Hunter → Gonzo → Devil's Advocate → Integrator.

Do I need to label the persona in the output?

Labeling is optional. Use labels when you want transparency about reasoning mode; omit them when you prefer a seamless, integrated response.