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 skills/ace3000chao/book2startup/topgradingnpx skills add ace3000chao/book2startup --skill topgradinggit clone --depth 1 https://github.com/ace3000chao/book2startupWhat 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 | $0.00109 | $0.02844 |
| Opus 5 | $0.00055 | $0.01422 |
| Sonnet 5 | $0.00022 | $0.00569 |
| Haiku 4.5 | $0.00011 | $0.00284 |
Grade A, and why
Topgrading（招聘A类人才系统方法论） 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 3d 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 — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Topgrading（招聘A类人才系统方法论）
R — 原文 (Reading)
"Topgrading is a methodology for hiring A Players, co-created by Brad and Geoff Smart. It has a proven track record of helping leaders hire the right person more than 90% of the time vs. the 25% to 60% success rates with feel-good conversations. The core tools: Job Scorecard to define success, sequenced behavioral interviews to assess candidates, and a reference process called TORC."
— Verne Harnish, Scaling Up, 第4章 The Team
中文翻译：Topgrading是一套招聘A类人才的方法论，由Brad和Geoff Smart共同创建。它帮助领导者将招聘成功率从"凭感觉谈话"的25%-60%提升到90%以上。核心工具包括：定义成功标准的Job Scorecard、评估候选人的序列行为面试，以及名为TORC的背景调查流程。
I — 方法论骨架 (Interpretation)
Topgrading是一套**让招聘从"赌博"变成"系统"**的方法论。它的核心洞察是：大多数招聘失败不是运气不好，而是方法错了——靠感觉面试、靠直觉判断，最终结果自然随机。
三大核心工具：
-
Job Scorecard（岗位记分卡）
- 在招聘前写好"成功的样子"：5-8个核心成果+对应的衡量标准
- 不是岗位职责清单，而是"6个月后这个人做出什么说明他成功了"
- 作用：招聘前对齐标准，避免"进来后才发现期望不一致"
-
序列行为面试（Sequenced Behavioral Interview）
- 每个面试官只问1-2个问题，但追问到行为细节（STAR格式）
- S/T = Situation/Task（当时什么情况/要完成什么）
- A = Action（你具体做了什么）
- R = Result（结果如何，量化）
- 追问"你"而非"我们"，区分真实经历和编造故事
- 序列设计：先非压力面试建立信任，再深入挖掘能力
-
TORC背景调查（Targeted Reference Check）
- 不是泛泛的"前雇主评价"，而是针对Scorecard设计的结构化提问
- 问前雇主："他在哪些方面做到了Scorecard的要求？"
- 问"他在哪些方面不符合你的期望？"（这是关键问题）
- 至少3个推荐人，拒绝提供则直接淘汰
关键心态：一个A类人才等于三个B/C类人才的生产力。不要因为"凑合用人"比"找不到A类"更省事而妥协。
A1 — 书中的应用 (Past Application)
案例 1: 杰克·韦尔奇重塑GE人才体系
- 问题: GE在韦尔奇时代早期，中层管理臃肿，B/C类人才占据大量岗位
- 方法论的使用: 韦尔奇推行"人才九宫格"，每年强制排序：20%明星（A类）、70%中坚（B类）、10%淘汰（C类）。配合Topgrading思维：不给C类人才找借口，直接要求提升或离开
- 结论: 人才密度是竞争优势——GE的强大不是因为某个明星，而是整体人才质量
- 结果: GE在韦尔奇时代成为全球最值钱的公司之一，文化延续数十年
案例 2: 科技创业公司的"感觉招聘"陷阱
- 问题: 一家SaaS公司创始团队技术很强，但招来的销售VP、运营总监纷纷在12-18个月内离职，创始人对"为什么总是看走眼"感到困惑
- 方法论的使用: 引入Job Scorecard——对销售VP定义"6个月内签约500万ARR"等可量化成果，而非"有经验、能带团队"。用序列行为面试重新评估候选人，发现之前招的人都在简历上看起来不错但缺乏真实案例
- 结论: 不是"看人眼光差"，是招聘流程本身有缺陷
- 结果: 新销售VP在12个月内签约680万ARR，留任超过3年
案例 3: 餐饮连锁的人才本土化
- 问题: 一家快速扩张的餐饮连锁向二三线城市拓展，发现本地招募的店长质量参差不齐，总部派驻成本高且不可持续
- 方法论的使用: 设计标准化的"店长Scorecard"：门店QSC评分、卫生合规、员工流失率、坪效等5项核心指标。招聘时用行为面试问"告诉我你曾经让一家门店从差变好的具体经历"，追问细节
- 结论: 二三线城市不缺A类人才，缺的是识别标准和方法
- 结果: 本土店长A类比例从30%提升至65%，外派成本降低70%
What ships with it
1 file 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.
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.
- 3d ago First seen · 174 lines · 109 tokens per session scan A f462c90f9c9e
Topgrading（招聘A类人才系统方法论） is a skill published in the GitHub repository ace3000chao/book2startup (72 stars, last pushed 4mo ago), licensed MIT. It adds 109 tokens to every session and 2,844 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
Applicant Screening
Screen job applications against requirements and score candidates.
interview-signal
Use when an interviewer needs an evidence-driven interview guide, a live follow-up probe, or a multi-round handoff based on a candidate resume, job description, interviewer perspective, and any previous interview questions and candidate responses.
headcount-and-comp-planning
Guide headcount planning and compensation strategy — hiring prioritization, compensation benchmarking, burn impact modeling, equity grant frameworks, and hiring ROI analysis for strategic talent investment.
justhireme-disabled
Deprecated placeholder. Do not use as project guidance.
Hiring Review
Use when reviewing an employment offer letter (and any restrictive-covenant exhibits) against the candidate's actual work jurisdiction and the employer's standard hiring practices to produce a structured draft review memo with a verdict, for attorney review.
mingle
Agent-powered professional networking inside your chat. Like LinkedIn, but your AI does the networking. Find collaborators, co-founders, freelancers, experts. Double opt-in, cryptographic trust, zero spam.