reasoning-personas

What This Skill Does

The reasoning-personas skill is a powerful cognitive framework for the OpenClaw AI agent, designed to shift the agent's internal heuristic processing. By activating specific behavioral personas, users can guide the agent to prioritize different analytical lenses, such as radical inquiry, defensive planning, pattern matching, or system integration. Rather than operating with a generic reasoning style, these personas adjust the agent's reward functions and question frameworks, ensuring that responses are tuned to the specific needs of the current task. This tool is essential for moving beyond surface-level answers and into deeper, structured, and multidimensional analysis.

Installation

To install this skill, use the following command in your OpenClaw environment: clawhub install openclaw/skills/skills/artyomx33/reasoning-personas

Once installed, the persona-based reasoning will automatically integrate with your existing workflows, such as code reviews and plan drafting, while remaining available for manual invocation.

Use Cases

• Strategic Brainstorming: Use the 'Gonzo Truth-Seeker' persona to identify hidden biases, challenge status-quo assumptions, and explore unconventional solutions when traditional ideation hits a wall.
• Risk Mitigation: Before finalizing high-stakes decisions or deploying code, invoke the 'Devil's Advocate' to stress-test your plan, expose failure modes, and identify potential weak points that might collapse under stress.
• System Architecture: Use the 'Pattern Hunter' and 'Integrator' personas to ensure that new features align with established precedents and maintain overall system coherence, preventing technical debt.
• Complex Problem Solving: Cycle through the full Multi-Persona Analysis—starting with the Pattern Hunter for context, moving to Gonzo for innovation, testing with the Devil's Advocate, and finishing with the Integrator for final coherence.

Example Prompts

1. "Put on your Gonzo hat; I need you to tell me why this business plan might fail even if we do everything right."
2. "Devil's advocate this: I'm planning to migrate our entire database to a new server next weekend. What could go wrong?"
3. "What precedents apply to this UI decision? I want to make sure we're not reinventing the wheel."

Tips & Limitations

• Auto-Activation: Be aware that the skill auto-activates during common tasks like code-review or brainstorming. You can override these defaults if you prefer a neutral tone.
• Compound Thinking: For the best results on critical projects, manually trigger the Multi-Persona Analysis. Running these sequentially is significantly more effective than asking for a single response.
• Output Markers: You can toggle the prefixing (e.g., [Gonzo mode]) by configuring your agent settings if you prefer the internal reasoning process to remain completely invisible.