Debate Tribunal — Development agent for Claude Code
Multi-agent structured debate protocol where 3-5 persona-driven debaters argue positions through claim-evidence-rebuttal rounds, moderated by a judge with adaptive convergence detection.
How to install Debate Tribunal
Installs to ~/.claude/agents/shaheerkhawaja-productionos-debate-tribunal.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/ShaheerKhawaja/ProductionOS/HEAD/agents/debate-tribunal.md -o ~/.claude/agents/shaheerkhawaja-productionos-debate-tribunal.md Restart Claude Code, or start a new session, for it to be picked up.
What Debate Tribunal does
name: debate-tribunal description: "Multi-agent structured debate protocol where 3-5 persona-driven debaters argue positions through claim-evidence-rebuttal rounds, moderated by a judge with adaptive convergence detection. Implements recursive debate within Claude Code's single-model constraint." color: orange model: opus tools:
- Read
- Glob
- Grep subagent_type: productionos:debate-tribunal stakes: high
Alternatives in Development
- Workflow Plan Checker — Validates plan quality with up to 3 revision rounds 530 ★
- Finding Checker — Blind adversarial checker — given ONLY a finding artifact and its evidence (never the author's reasoning), tri 348 ★
- Fg Teaching Specialist — Finance Guru Teaching & Enablement Mentor (Maya Brooks) 317 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Op Debater
Read-only debate seat — holds ONE position across rounds, engages rival positions on their strongest form, and
Review Judge
Neutral zero-context judge for the adversarial-review skill. Rules ONLY on findings the author disputed — read
Convergence Monitor
Cross-iteration convergence tracker — runs 6 detection algorithms (score-based, semantic, diminishing returns,
Mockup Coordinator
INLINE playbook for the operator-driven mockup loop: dispatches mockup-designer and mockup-reviewer rounds unt
Capital Allocation
Capital Allocation Committee — CFO-led debate protocol for material capital decisions; growth-vs-discipline ad
Auditor Agent
Self-improving auditor. Reads persona evaluation traces and proposes bounded, evidence-driven optimizations to
Related Skills
AI Debate Club
A multi-agent debate engine. Two different LLMs argue any motion, research the web for themselves, and cite th
Spar
Make the models argue before you build — Codex attacks the plan in bounded read-only rounds, then one builds a
Claude Council Skill
A Claude skill that convenes 7 expert AI personas to debate any decision, idea, or problem — then synthesizes