Borrowing it
Nothing to install: this file belongs to MitsudoAI/guru-pk-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/MitsudoAI/guru-pk-mcp/main/.claude/commands/guru-pk.mdgit clone --depth 1 https://github.com/MitsudoAI/guru-pk-mcpWrote 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/mitsudoai/guru-pk-mcp/guru-pk)<a href="https://agentmods.dev/commands/mitsudoai/guru-pk-mcp/guru-pk"><img src="https://agentmods.dev/badge/commands/mitsudoai/guru-pk-mcp/guru-pk/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/mitsudoai/guru-pk-mcp/guru-pk"><img src="https://agentmods.dev/badge/commands/mitsudoai/guru-pk-mcp/guru-pk.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00000 | $0.05345 |
| Opus 5 | $0.00000 | $0.02672 |
| Sonnet 5 | $0.00000 | $0.01069 |
| Haiku 4.5 | $0.00000 | $0.00534 |
Grade A, and why
guru-pk 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 11d 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 — 672 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guru-PK Expert Debate System
核心理念：真实人物 + 智能轮次 + Markdown输出 + 多语言支持
语言模板定义
根据检测到的语言，使用对应模板：
中文模板 (zh)
title: "# Guru-PK 专家辩论："
session_id: "会话ID"
created_time: "创建时间"
question: "问题"
debate_rounds: "辩论轮次"
participants: "参与专家"
identity: "身份"
core_views: "核心观点"
thinking_style: "思维特征"
round_1: "第一轮：独立思考"
round_2: "第二轮：交叉辩论"
round_3: "第三轮：最终立场"
round_4: "第四轮：智慧综合"
debate_summary: "辩论总结"
completed_time: "完成时间"
total_rounds: "总轮次"
total_words: "总字数"
mode_2rounds: "2轮（独立思考 + 交叉辩论）"
mode_3rounds: "3轮（独立思考 + 交叉辩论 + 最终立场）"
mode_4rounds: "4轮（独立思考 + 交叉辩论 + 最终立场 + 智慧综合）"
tip_default: "默认执行2轮（核心价值最高）"
tip_full: "如需四轮完整分析，可在问题中加入"完整辩论""
英文模板 (en)
title: "# Guru-PK Expert Debate: "
session_id: "Session ID"
created_time: "Created"
question: "Question"
debate_rounds: "Debate Rounds"
participants: "Participating Experts"
identity: "Identity"
core_views: "Core Views"
thinking_style: "Thinking Style"
round_1: "Round 1: Independent Thinking"
round_2: "Round 2: Cross Debate"
round_3: "Round 3: Final Position"
round_4: "Round 4: Wisdom Synthesis"
debate_summary: "Debate Summary"
completed_time: "Completed"
total_rounds: "Total Rounds"
total_words: "Total Words"
mode_2rounds: "2 Rounds (Independent Thinking + Cross Debate)"
mode_3rounds: "3 Rounds (Independent Thinking + Cross Debate + Final Position)"
mode_4rounds: "4 Rounds (Independent Thinking + Cross Debate + Final Position + Wisdom Synthesis)"
tip_default: "Default: 2 rounds (highest value)"
tip_full: "For full 4-round analysis, add 'complete debate' or 'full analysis' in question"
日语模板 (ja)
title: "# Guru-PK 専門家討論："
session_id: "セッションID"
created_time: "作成日時"
question: "質問"
debate_rounds: "討論ラウンド"
participants: "参加専門家"
identity: "身分"
core_views: "中心的見解"
thinking_style: "思考特性"
round_1: "第1ラウンド：独立思考"
round_2: "第2ラウンド：クロス討論"
round_3: "第3ラウンド：最終立場"
round_4: "第4ラウンド：知恵の統合"
debate_summary: "討論まとめ"
completed_time: "完了時刻"
total_rounds: "合計ラウンド"
total_words: "総文字数"
mode_2rounds: "2ラウンド（独立思考 + クロス討論）"
mode_3rounds: "3ラウンド（独立思考 + クロス討論 + 最終立場）"
mode_4rounds: "4ラウンド（独立思考 + クロス討論 + 最終立場 + 知恵の統合）"
tip_default: "デフォルト：2ラウンド（最高価値）"
tip_full: "完全な4ラウンド分析が必要な場合、質問に「完全討論」を追加してください"
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.
- 11d ago First seen · 672 lines · 0 tokens per session scan A 36f04938bf20
guru-pk is a command published in the GitHub repository MitsudoAI/guru-pk-mcp (6 stars, last pushed 11mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 5,345 tokens. 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
review-epo-claims
Analyze patent claims for EPO Art. 84 EPC compliance - clarity, conciseness, support by description.
ingest-dev
Ingest a URL, directory, or file into your knowledge base. remember = a specific durable fact, ingest = a URL, learn = a distilled lesson that gets retrieval preference.
quarry
Manage your quarry knowledge base.
remember-dev
Remember inline text content in your knowledge base. remember = a specific durable fact, ingest = a URL, learn = a distilled lesson that gets retrieval preference.
resume
Auto-detect and resume any interrupted Plan Cascade task. Detects mega-plan, hybrid-worktree, or hybrid-auto context and routes to the appropriate resume command.
scout
Scout vetted GitHub candidates for a thin shelf (or a stale entry's successor), deduped against the library, ending in /add-entry offers.