society-of-mind
$
npx mdskill add yogsoth-ai/de-anthropocentric-research-engine/society-of-mindFacilitates multi-agent collaborative debate to refine perspectives iteratively
- Solves complex problems by simulating diverse viewpoints and resolving disagreements
- Uses subagents like debate-architect, perspective-critic, and divergence-detection
- Decides based on iterative sharing and revision until convergence or divergence
- Delivers a synthesized verdict through debate-transcript-analysis
SKILL.md
.github/skills/society-of-mindView on GitHub ↗
---
name: society-of-mind
description: "Strategy: Multi-agent collaborative debate based on Du et al. Society of Mind. Agents share perspectives iteratively until convergence or divergence is detected."
type: strategy
used-by: [multiagent-debate]
tactics: [perspective-rotation]
---
# Society of Mind Strategy
Multiple agents evaluate independently, then share and revise through iterative rounds.
## Method
1. **debate-architect** assigns N distinct perspectives to agents
2. Each **perspective-critic** evaluates artifact from assigned viewpoint
3. All perspectives shared — agents revise positions given others' arguments
4. **divergence-detection** identifies agreement clusters and persistent disagreements
5. Repeat sharing rounds until convergence or saturation
## Budget Table
| Parameter | S | M | L |
|---|---|---|---|
| Debate rounds | 4 | 8 | 12 |
| Participating agents | 3 | 5 | 8 |
| Coverage dimensions | 3 | 5 | 7 |
| External evidence searches | 2 | 5 | 10 |
## Orchestration
```
debate-architect → [assign perspectives]
→ [parallel]: perspective-critic × N
→ [for each sharing round]:
share all outputs → perspective-critic revises
→ divergence-detection → (converge or continue)
→ debate-transcript-analysis → verdict-synthesis
```
## Subagents
- debate-architect (perspective assignment)
- perspective-critic × N (perspective evaluation)
- divergence-detection (convergence tracking)
- confidence-calibration (termination decision)
More from yogsoth-ai/de-anthropocentric-research-engine
- abductive-hypothesis-generationStrategy: 面对异常的最佳解释推理
- ablation-brainstormRemove components one by one, observe system changes to reveal hidden dependencies and generate ideas from structural gaps.
- ablation-component-mappingMap system architecture to ablatable units for ablation studies
- ablation-designDesign ablation studies to isolate component contributions in ML systems
- ablation-executionRemove components one by one from a system, record the response/impact of each removal.
- abp-vulnerability-classificationClassify assumptions on 2 axes — load-bearing (how much conclusion depends on it) × vulnerable (how likely to be false). Focuses attention on High-Load × High-Vulnerable quadrant.
- abstraction-extractionExtract abstract principles from concrete domain cases. Strips domain-specific details to reveal transferable mechanisms.
- abstraction-ladderPerform bisociation at multiple abstraction levels
- abstraction-ladderingMove between concrete and abstract framings — 3 levels up (Why?) and 3 levels down (How?) to find the most productive research level.
- abstraction-to-designAbstract biological principle to design principle. Bridge from biology to engineering.