Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/meesp123/claude-code-agent-team-debate/debategit clone --depth 1 https://github.com/meesp123/claude-code-agent-team-debateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/meesp123/claude-code-agent-team-debate/debate)<a href="https://agentmods.dev/commands/meesp123/claude-code-agent-team-debate/debate"><img src="https://agentmods.dev/badge/commands/meesp123/claude-code-agent-team-debate/debate.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00021 | $0.10328 |
| Opus 5 | $0.00010 | $0.05164 |
| Sonnet 5 | $0.00004 | $0.02066 |
| Haiku 4.5 | $0.00002 | $0.01033 |
Grade A, and why
debate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 860 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/debate — Agent Team 辩论赛
目的：多个 AI agent 互相辩论，Lead 做 moderator 汇总 DECISION。 原则：agents 互相发消息讨论，不是线性流水线。外部 AI (Pi/Codex/Gemini/Qwen/Kimi) 通过 proxy agent 参与。
Phase Map
| # | Phase | 依赖 | 可跳过 |
|---|---|---|---|
| 1 | 参数解析 + Context Assembly | - | ❌ |
| 2 | Team Setup (创建 + Spawn) | 1 | ❌ |
| 2.5 | Wait for READY | 2 | ❌ |
| 3 | Round 1: PROPOSAL | 2.5 | ❌ |
| 4 | Round 2: CHALLENGE | 3 | ❌ |
| 4.9 | Context Refresh | 4 | ✅ (仅 rounds>=3) |
| 5 | Round 3: REBUTTAL (rounds>=3) | 4.9 | ✅ |
| 6 | DECISION + Cleanup | 4/5 | ❌ |
Phase 1: 参数解析 + Context Assembly
从 $ARGUMENTS 中解析：
- topic: 除
--flags外的所有文本 - files:
--files后的文件路径列表（可选） - with:
--with后逗号分隔的外部 CLI 列表（可选，支持 pi/codex/gemini/qwen/kimi） - rounds:
--rounds后的数字（默认 2，最多 3） - preset:
--preset后的预设名（可选，支持 balanced/deep，默认 balanced） - model:
--model后逗号分隔的cli:model对（可选，如pi:gpt-5.2,codex:gpt-5.3-codex,gemini:gemini-3-pro-preview） - effort:
--effort后的 reasoning effort 级别（可选，对 pi/codex/kimi 有效。pi 用--thinking level，codex 用--config，kimi 用--thinkingtoggle。支持 high/xhigh） - advocate_model:
--advocate后的 Claude 模型（可选，支持 haiku/sonnet/opus） - critic_model:
--critic后的 Claude 模型（可选，支持 haiku/sonnet/opus）
Preset 智能默认
Preset 提供基线配置，显式 flag (--advocate/--critic/--model/--effort) 覆盖 preset：
| Preset | advocate | critic | pi | pi effort | codex | codex effort | gemini | qwen | kimi | kimi thinking |
|---|---|---|---|---|---|---|---|---|---|---|
balanced (默认) |
sonnet | sonnet | — | high | — | high | — | — | — | off |
deep |
opus | sonnet | gpt-5.2 | xhigh | gpt-5.2 | xhigh | gemini-3-pro-preview | — | — | on |
— = 不传
-Model，使用 CLI 自身默认模型（pi → gpt-5.3-codex, codex → gpt-5.3-codex, gemini → gemini-2.5-pro, qwen → coder-model） 显式--model覆盖 preset 的模型，如--preset deep --model gemini:gemini-3-flash-preview只覆盖 gemini
可用模型参考
Lead spawn proxy agent 时按此表选择 -Model 参数：
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 860 lines · 21 tokens per session scan A 4068b7726427
debate is a command published in the GitHub repository meesp123/claude-code-agent-team-debate (2 stars, last pushed 6mo ago), licensed MIT. It adds 21 tokens to every session and 10,328 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
edison.list
This obsolete slash command is disabled for this repository.
edison.plan-mode
This obsolete slash command is disabled for this repository.
edison.qa-audit
This obsolete slash command is disabled for this repository.
edison.qa-bundle
This obsolete slash command is disabled for this repository.
edison.qa-new
This obsolete slash command is disabled for this repository.
edison.qa-promote
This obsolete slash command is disabled for this repository.