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 skills add PANGKAIFENG/ai-product-manager-skills --skill ai-collaboration-calibrationgit clone --depth 1 https://github.com/PANGKAIFENG/ai-product-manager-skillsWrote 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/skills/pangkaifeng/ai-product-manager-skills/ai-collaboration-calibration)<a href="https://agentmods.dev/skills/pangkaifeng/ai-product-manager-skills/ai-collaboration-calibration"><img src="https://agentmods.dev/badge/skills/pangkaifeng/ai-product-manager-skills/ai-collaboration-calibration/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/skills/pangkaifeng/ai-product-manager-skills/ai-collaboration-calibration"><img src="https://agentmods.dev/badge/skills/pangkaifeng/ai-product-manager-skills/ai-collaboration-calibration.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 | $0.00272 | $0.02159 |
| Opus 5 | $0.00136 | $0.01079 |
| Sonnet 5 | $0.00054 | $0.00432 |
| Haiku 4.5 | $0.00027 | $0.00216 |
Grade A, and why
ai-collaboration-calibration 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 13d 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI 协作校准 Skill
中文速查
- 中文名：协作校准 / 认知校准 / 问题脑暴
- 英文稳定名：
ai-collaboration-calibration - 分类：认知与协作
- 你可以这样叫我：
帮我想想、先聊一下、一起脑暴、先别执行，帮我看清问题、挑战我的假设、这个方案是不是想错了、帮我做认知校准、双向钢人、钢人论证一下、正反最强论证 - 适合：问题还没定义清楚时的脑暴发散，以及已有问题框架但目标、领域、约束或判断标准可能偏了时的结构化校准
- 不适合：翻译、摘要、格式转换、明确的小改动；已确认问题上的成熟方案压力测试应改用
grill-me
核心行为：反转 AI 的默认模式——从「顺着用户补充」变为「先挑战假设和问题定义」。
反模式识别和核心原则：加载 references/anti-patterns.md
第一步：判断协作层级
收到输入后，先判断层级，再决定走哪条路径。
| 层级 | 模式 | 典型诉求 | 行动 |
|---|---|---|---|
| L1 | 执行器 | 「帮我做」：写文案、整理表格 | 直接执行 |
| L2 | 优化器 | 「帮我改好」：优化结构、润色 | 直接优化 |
| L3 | 挑战者 | 「帮我找错」：反驳问题定义、目标和约束假设 | 进入挑战模式；带着初步想法要快速正反验证并裁决时用 13-steelman-verdict；若已有具体方案且问题已确认，转 grill-me |
| L4-fuzzy | 脑暴模式 | 问题极度模糊、描述的是感受/现象/解法 | 进入脑暴路径 |
| L4-framed | 校准模式 | 问题有雏形但方案越补越复杂 | 进入 6 步校准 |
L4-fuzzy 信号：描述的是解法/功能/感受、对话处于最早期、说「帮我想想」「先聊一下」 L4-framed 信号：已有问题框架但反复改方案、问「这个方向对不对」
L4-fuzzy：脑暴路径
不走 6 步结构化流程。并行激活三个动作，在问题定义完成前不输出任何方案候选。
动作 1 — JTBD 追问：描述的是功能/渠道时，问「如果这个成功了，真正完成的是什么任务？」 动作 2 — 假设显化：列出隐含前提让用户确认哪个最不确定 动作 3 — 说出判断：每 3-4 轮说一次「我的判断是 X 而不是 Y，原因是 Z——你认同吗？」
Done Signal（AI 主动触发三问）→ 详见 references/brainstorm-mode.md
L4-framed：6 步校准
不是每步都必须完整跑，答不上来说明还没想清楚。
| 步骤 | 目标 | 详细模板 |
|---|---|---|
| Step 1 问题分类 | 判断是已知解/设计/发现/棘手/组织/数据治理问题 | references/modes/05-problem-classify.md |
| Step 2 提升层级 | 向上抽象 3 层，判断精力是否投在正确层级 | references/modes/11-level-up.md |
| Step 3 领域定位 | 找到问题在成熟领域里的名字和标准解 | references/modes/02-domain-mapping.md |
| Step 4 最佳实践 | 先建立「不考虑约束的标准解」参照系 | references/modes/04-best-practice.md |
| Step 5 裂缝定位 | 找到解决后其他复杂度会坍塌的那个 Crux | references/modes/08-crux.md |
| Step 6 挑战假设 | 指出最可能错误的 2-3 个假设及验证方式 | references/modes/01-challenge.md |
输出格式：加载 references/output-format.md
追加模式路由
| 信号 | 调用模式 |
|---|---|
| 不知道问题属于哪个领域 | 09-expert-role + 02-domain-mapping |
| 「哪里不对劲」但说不清 | 03-blind-spot |
| 感觉被约束卡死 | 07-solution-space + 04-best-practice |
| 已有问题框架，想提前识别问题定义层面的失败风险 | 06-failure-premortem |
| 问题未确认，但已有初步想法，要正反最强论证并拿到明确裁决 | 13-steelman-verdict |
| 问题已确认，已有具体方案 / 架构 / 决策，想做压力测试 | 转交 grill-me |
| 方案确定，想规划演进 | 10-upgrade-path |
| 讨论细节很久无进展 | 11-level-up |
| 探索结束，想沉淀资产 | 12-asset-capture |
What ships with it
21 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- evals/check-rules.md 2.2 KB
- evals/evals.json 4.0 KB
- evals/test-prompts.csv 1.6 KB
- references/anti-patterns.md 3.1 KB
- references/brainstorm-mode.md 4.5 KB
- references/modes/01-challenge.md 1.1 KB
- references/modes/02-domain-mapping.md 1.3 KB
- references/modes/03-blind-spot.md 1.0 KB
- references/modes/04-best-practice.md 1.1 KB
- references/modes/05-problem-classify.md 1.7 KB
- references/modes/06-failure-premortem.md 1.4 KB
- references/modes/07-solution-space.md 1.3 KB
- references/modes/08-crux.md 1.3 KB
- references/modes/09-expert-role.md 1.1 KB
- references/modes/10-upgrade-path.md 1.1 KB
- references/modes/11-level-up.md 1.2 KB
- references/modes/12-asset-capture.md 1.6 KB
- references/modes/13-steelman-verdict.md 2.6 KB
- references/output-format.md 1.2 KB
- references/quality-checklist.md 1.9 KB
- references/standard-opening.md 1.8 KB
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.
- 13d ago First seen · 126 lines · 272 tokens per session scan A f1b8f7ad9183
ai-collaboration-calibration is a skill published in the GitHub repository PANGKAIFENG/ai-product-manager-skills (11 stars, last pushed 14d ago), licensed MIT. It adds 272 tokens to every session and 2,159 once invoked, about $0.0014 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 skills, from other repositories
baoyu-youtube-transcript
A tool for downloading the written captions, subtitles, chapter information, speaker labels, and cover image from a YouTube video using its URL or ID.
baoyu-comic
A tool for creating original educational comics that explain knowledge or ideas through multiple illustrated panels. It supports different art styles and tones and can create several comics in one batch.
baoyu-diagram
Create professional, dark-themed SVG diagrams of any type — architecture diagrams, flowcharts, sequence diagrams, structural diagrams, mind maps, timelines, illustrative/conceptual diagrams, and more. Use this skill whenever the user asks for any kind of technical or conceptual diagram, visualization of a system…
playwright-cli
Automates browser interactions for testing and validating your own web applications using playwright-cli. Use when you need terminal-first browser control for navigation, form filling, screenshots, tracing, bound browser sessions, debugging, or generating Playwright test code. Only use against applications you own or…
agent-wiki
Incremental LLM-friendly wiki generator for Obsidian note vaults. Use when: (1) Building wiki from notes, (2) Ingesting notes to wiki, (3) Obsidian LLM wiki, (4) Incremental knowledge base management. Triggers: 'build wiki from notes', 'ingest notes to wiki', 'Obsidian LLM wiki', 'incremental knowledge base'.
context-optimizer
A context-management assistant for long conversations with AI agents. It helps organize the information an agent needs when conversations become large or contain conflicting details.