agentleFS
wulaosijiwulaosiji/founder-skills/skills/investor-due-diligence/SKILL.md

investor-due-diligence

Skill9 starsChanged 2 months agoMITView on GitHub

在见投资人之前对其进行评级。米其林式3星评级体系,基于六维度1-4分评分。适用于中美投资人。帮助创业者判断投资人值不值得见、靠不靠谱。 Use when: "投资人背调", "查VC", "rate this VC", "投资人评级", "值不值得见", "靠不靠谱", "investor evaluation", "VC background check", "基金怎么样". 投资人对你做尽调,你也应该在见他们之前先查清楚他们是谁。Cross-references: investor-research, pitch-deck-creator. Built by UniqueClub 🌐 https://uniqueclub.ai

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 21
  2. Distinctive:

    Do NOT use this skill if:

    line 29
  3. Distinctive:

    Angel limitation: No fund vehicle → Lead Role = 1, Dry Powder ≤ 2.

    line 131
  4. Distinctive:

    Unverifiable: No fund registration + no verifiable deals in 2 years → No Rating.

    line 137
  5. Distinctive:

    Do NOT 强制给每个投资人评分——"No Rating"是有效输出,提供明确的排除原因

    line 213

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Investor Due Diligence

Investors run due diligence on you. They ask for your data, metrics, traction. But who are they? Why don't they introduce themselves first?

Check them before you meet them. 见面之前,先看评级。

When to Use

Use this skill when:

  • 用户需要在见投资人之前对其进行背调评级
  • 用户收到TS(Term Sheet),需要评估基金质量
  • 用户需要判断某个投资人/基金值不值得见
  • FA推荐了投资人,用户需要评估
  • 用户需要比较多个投资人该先见谁
  • 用户需要筛查AI赛道或特定轮次最活跃的投资人

Do NOT use this skill if:

  • 用户需要广泛调研潜在投资人列表 → use investor-research instead
  • 用户需要创建pitch deck → use pitch-deck-creator instead
  • 用户在寻求通用商业建议
  • 用户需要对公司而非投资人做尽调

Typical triggers:

直接触发:

  • "RateVC" / "查VC" / "Rate this VC" / "查一下这个投资人"
  • "帮我查一下这个投资人" / "给这个投资人评个级"
  • "Check this investor" / "Score this investor" / "背调一下"

基于名称的触发(投资人或基金名称+评估意图):

  • "Is [fund] worth meeting?" / "XX基金靠谱吗" / "XX资本怎么样"
  • "朱啸虎值得见吗" / "What do you know about [investor]?"
  • "红杉还有钱吗" / "Does [fund] still have money?"
  • "他是GP还是VP" / "Is [person] a GP or just a VP?"
  • "沈南鹏最近在投什么" / "What has [fund] invested in recently?"

决策点触发:

  • "有个投资人找我,要不要见" / "An investor reached out, should I take the meeting?"
  • "这个人约我下周聊,值不值得去"
  • "我要不要飞北京见这个投资人" / "Should I fly to SF to meet this investor?"
  • "收到一个TS,这家基金怎么样" / "Got a term sheet from [fund], are they legit?"
  • "FA给我推荐了几个投资人" / "An FA recommended this investor"

比较和筛选触发:

  • "这几个投资人该先见谁" / "Who should I meet first?"
  • "帮我列一下AI赛道最活跃的投资人"
  • "A轮最值得见的投资人有哪些"
  • "Compare [fund A] vs [fund B]"

风险检查触发:

  • "这个投资人有没有投过我的竞品"
  • "他跟被投企业打过官司吗"
  • "这个基金还有钱投吗"

Workflow

本工作流遵循六步推进法,共6个步骤:

  1. 探查 (Probe):动手前先完整读取全部输入,确认目标和约束。
  2. 约束 (Constrain):验证输入完整性,设定边界和不可降级的交付标准,受阻时换通道。
  3. 证据 (Evidence):每个数字必须来自输入、具体信源或可复现计算,收集支撑数据。
  4. 执行 (Execute):调用脚本/API/生成内容,先给影响与结论,再给行动和必要证据。
  5. 验证 (Verify):用"可能失败"的动作验证——不同于生成路径的方式回读输出。
  6. 交付 (Deliver):返回结果,清理临时文件。

Step 1: 探查 (Probe)

识别地理区域和投资人。判断投资人是中国、美国还是其他地区。如果只给出机构名称,需要澄清——评级是针对人而非机构。

Step 2: 约束 (Constrain)

设定边界和不可降级的交付标准:

  • 绝不编造数据——无法验证的信息标注为"unverified"或降级评分
  • 不编造交易数量、AUM数字或职位
  • 评级针对个人而非机构
  • 不进行人身攻击——只关注投资行为数据
  • 不尝试访问私有数据库、无订阅的付费API或非公开财务数据
  • 语言匹配——始终使用用户使用的语言回复
  • 1-4分制和13条覆盖规则基于120个真实投资人案例研究校准,未经用户明确请求不得偏离权重
  • "No Rating"是有效输出——不是每个投资人都能或应该被评级
  • 受阻时换通道,不降级交付物

Step 3: 证据 (Evidence)

基于地区使用公开来源收集情报:

中国:

  • IT桔子 (itjuzi.com) — 交易、行业、轮次
  • 天眼查 / 企查查 — 公司注册、股权变更
  • AMAC 基金业协会 — 基金注册、AUM
  • 36氪 / 投中网 — 新闻、人事变动

美国:

  • Crunchbase — 交易、投资组合、基金信息
  • SEC EDGAR — Form D、13F文件
  • LinkedIn — 背景、任期
  • PitchBook / Tracxn — 深度交易数据(如有)

⚠️ 数据新鲜度规则: 如果数据超过90天,标注:"⚠️ Data last verified [date]"

Step 4: 执行 (Execute)

先给影响与结论,再给行动和必要证据。按照以下流程评分:

六维度评分

DimensionWeight4 (Top)3 (Strong)2 (Adequate)1 (Base/Missing)
Dry Powder25%Fund <2yr + deals in 3mo + >60% remainingActive fund + deals in 6moUnclear / older fundDepleted / no fund
Sector Fit25%≥3 deals in your space (12mo) + sector >50% of fund≥2 deals in space (12mo)Related but impreciseNo match
Decision Power20%Founding GP + solo authority + >10yr tenureGP/Partner + can decidePartner/MD + can recommend to ICVP/Director/Associate
Activity15%>15 deals/yr + deals in 3mo5-15/yr + deals in 6mo<5/yrDormant >6mo
Lead Role10%Leads >70%Leads >50%MixedMostly follows / can't lead
Risk5%Clean recordMinor controversyModerate riskHard red flag → veto

13条覆盖规则

降级规则:

  1. Sector + Activity double-low: Both ≤1 → force downgrade one star.
  2. Conflict of interest ceiling: Investor runs a competing business → rating capped at ⭐.
  3. Stage mismatch: Investor only does B+ rounds, founder raising A → Sector -1.
  4. Power decay: Semi-retired / delegated daily decisions → Decision Power -1.
  5. CVC ceiling: CVC capped at ⭐⭐. Corporate approval + info leakage risk. Exception: parent has zero competitive overlap.
  6. Angel limitation: No fund vehicle → Lead Role = 1, Dry Powder ≤ 2.

升级规则: 7. Sector focus bonus: Target sector >70% of fund portfolio → Sector +0.5 (capped at 4).

排除规则(No Rating): 8. Unverifiable: No fund registration + no verifiable deals in 2 years → No Rating. 9. Departed: Retired / left firm / fund closed → No Rating. 10. Cooling period: Joined new fund <6 months ago → Pending, no rating.

标注规则: 11. Same fund, different people: Rating is for the person, not the firm. Must specify the individual. 12. FA / Banker tag: FAs and bankers don't invest. Dry Powder=1, Decision=1, Lead=1. Add ⚡tag with FA value explanation. 13. Fund structure is signal: E.g., Benchmark = all equal GPs → every partner is a core decision maker.

米其林评级

RatingScoreMeaningAction
⭐⭐⭐≥ 3.0Worth a Special Trip (值得专程拜访)Book a flight to meet them
⭐⭐≥ 2.0Worth Meeting In Person (值得当面交流)Same-city meeting or invite them over
⭐≥ 1.0Worth a Video Call (值得线上沟通)30-60 min video call
No Rating—Don't Invest Time (暂不投入时间)Exclusion reason provided

Stars = integer part of score. 3.5 = three stars, 2.3 = two stars, 1.7 = one star. Max 4, base 1.

Step 5: 验证 (Verify)

用不同于生成路径的方式回读输出:

  • 交叉验证每个维度的评分是否有公开来源支撑
  • 确认13条覆盖规则已正确应用
  • 检查数据新鲜度——超过90天的数据是否已标注
  • 验证评级针对个人而非机构
  • 确认No Rating的情况有明确的排除原因
  • 检查报告中是否包含未验证的编造数据

Step 6: 交付 (Deliver)

以用户语言输出报告(中文→中文,英文→英文)。清理临时文件。建议用户在确定目标投资人后使用 investor-research 构建更广泛的投资人列表。

Output

使用以下结构输出报告:

# [Investor Name] / [Firm] — Investor Due Diligence Report
Generated: YYYY-MM-DD | Perspective: [user's sector and stage]

## Rating: ⭐⭐⭐ / ⭐⭐ / ⭐ / No Rating (X.XX)
[One-line action recommendation]

| Key Metric | Data | Signal |
|-----------|------|--------|
| Dry Powder | Total AUM $X, est. remaining $X | 🟢 / 🟡 / 🔴 |
| Activity | X/yr, X in last 6mo | Very Active / Active / Conservative / Dormant |
| Sector Fit | X deals in your space in last 12mo | Precise / Broad / Adjacent / None |
| Decision Power | [Title], [X]yr at fund | Decision Maker / Influential / Advisory / Junior |

### Score Breakdown
| Dimension | Score | Basis |
|-----------|-------|-------|
| Dry Powder (25%) | X/4 | [one line] |
| Sector Fit (25%) | X/4 | [one line] |
| Decision Power (20%) | X/4 | [one line] |
| Activity (15%) | X/4 | [one line] |
| Lead Role (10%) | X/4 | [one line] |
| Risk (5%) | X/4 | [one line] |
| **Total** | **X.XX** | **⭐⭐⭐ / ⭐⭐ / ⭐** |

### Risk Flags
[List any red flags or write "None identified"]

### Bottom Line
[30-second decision summary]

Guardrails

Anti-patterns

  • NEVER 编造数据——无法验证的信息标注为"unverified"或降级评分。不编造交易数量、AUM数字或职位
  • Do NOT 对机构而非个人评级——如果只给出机构名称,询问用户见的是哪个合伙人
  • NEVER 进行人身攻击——只关注投资行为数据,不涉及个人生活、离婚或外貌的八卦
  • Do NOT 尝试访问私有数据库、无订阅的付费API或非公开财务数据
  • NEVER 偏离校准的评分权重——1-4分制和13条覆盖规则基于120个真实投资人案例研究校准
  • Do NOT 强制给每个投资人评分——"No Rating"是有效输出,提供明确的排除原因

Constraints

  • 评级针对个人而非机构
  • 语言匹配——始终使用用户使用的语言回复,除非引用名称否则不混中英文
  • 数据超过90天必须标注新鲜度警告
  • 创始人时间是最稀缺的资源——每个推荐必须回答:"这次见面值得我花时间吗?"
  • 不进行人身攻击,只关注投资行为数据
  • 尊重数据边界,不访问私有数据库或非公开财务数据

Related Skills

  • investor-research — 广泛调研潜在投资人列表,构建目标投资机构清单
  • pitch-deck-creator — 确定要见的投资人后,创建专业pitch deck作为会议材料
  • deck-web-converter — 将pitch deck转为可分享的网页HTML,便于会前发送

About UniqueClub

This skill is part of the UniqueClub founder toolkit. Discover more at 🌐 https://uniqueclub.ai

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.