Back to list
stephenwinters81

triumvirate

by stephenwinters81

0🍴 0📅 Jan 13, 2026

SKILL.md


name: triumvirate description: 3x3 Parallel Debate system for high-confidence research. Spawns 9 agent instances (3 Proposer/Skeptic/Judge cohorts) to argue, critique, and adjudicate claims. Use for complex queries, controversial topics, or when hallucination-free research is critical.

The Triumvirate: Truth-Seeking Research Protocol

When to Activate

Use this skill when the user requests:

  • Research on complex or contested topics
  • High-confidence fact verification
  • Analysis where hallucinations are unacceptable
  • Scientific or technical claims that need rigorous validation
  • User explicitly mentions "triumvirate" or "debate analysis"

Architecture

9 Agent Instances organized as:

The Cohort (x3 parallel units)

  1. Proposer - Builds the strongest affirmative case with evidence
  2. Skeptic - Attacks the argument for logical fallacies, assumptions, overgeneralizations
  3. Judge - Impartially determines which claims are reality-compliant

The Synthesis

A final meta-analysis compares all three Judge verdicts to identify:

  • High Confidence Findings (all cohorts agree)
  • Areas of Uncertainty (cohorts disagree)
  • Refuted/Hallucinated Claims (caught by Skeptics)

Invocation

python3 ~/.claude/skills/triumvirate/triumvirate.py "Your research query here"

Example Usage

User: "What is the actual efficacy of intermittent fasting for weight loss?"

python3 ~/.claude/skills/triumvirate/triumvirate.py "What is the actual efficacy of intermittent fasting for weight loss?"

Why This Works

  1. Skeptic Isolation: The Skeptic is prompted as an "adversary," breaking sycophancy loops
  2. Condorcet Jury Theorem: 3 independent cohorts act as error-correctors during synthesis
  3. Logical Fallacy Detection: Skeptics explicitly check for correlation/causation errors, cherry-picking, survivorship bias, etc.

Requirements

  • Claude Code CLI installed and authenticated (claude command available)
  • Python package: rich (for formatted output)

Score

Total Score

50/100

Based on repository quality metrics

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

Reviews

💬

Reviews coming soon