private-company-research

Researches private companies using six parallel agents for valuation and risk analysis.

16.4k|2.5k|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/xbtlin/ai-berkshire --skill private-company-research-xbtlin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: private-company-research
Source: https://github.com/xbtlin/ai-berkshire/tree/main/codex-skills/private-company-research
Command: npx skills add https://github.com/xbtlin/ai-berkshire --skill private-company-research-xbtlin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Private companies like Ant Group, Xiaohongshu, SpaceX, or Stripe lack standardized financial disclosures, making it hard to judge what the business is actually worth. This Skill coordinates a multi-agent research team to piece together fragmented information, cross-validate data, and produce an investment-grade report with explicit confidence labels instead of vague both-sides analysis. ## Core Features & Use Cases - Six Parallel Research Agents: Business model decoding, financial data reconstruction and valuation, competitive landscape mapping, risk and governance assessment, technology and IP analysis, and alternative data signal mining (hiring, patents, litigation, app metrics). - Multi-Method Valuation: Combines latest funding round adjustment, comparable public companies, DCF scenarios, terminal value back-calculation, and transaction benchmarking into a weighted fair-value range with safety margin. - Anti-Bias Controls: Every data point carries a source, date, and confidence rating (high/medium/low); cross-agent signal consistency checks catch contradictions before the final verdict of invest, watch, or avoid. - Use Case: Ask it to research Xiaohongshu before a secondary share purchase, and receive a full report with estimated revenue, valuation range versus the latest round, exit path analysis, and a one-page decision table. ## Quick Start Ask the agent to run a private company deep research report on SpaceX covering business model, valuation, competition, risks, technology, and alternative data signals.

Frequently Asked Questions about private-company-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I research a private company with no public financials?

This Skill reconstructs financials from fragmented sources such as IPO prospectuses, parent company filings, bond documents, funding news, and industry reports. Each data point is labeled with its source, date, and a high, medium, or low confidence rating.

How to estimate the valuation of a pre-IPO startup?

The framework applies five methods: latest funding round adjusted for liquidation preferences, comparable public company multiples with liquidity discounts, three-scenario DCF, terminal value back-calculation, and recent transaction benchmarking. Results are weighted into conservative, fair, and optimistic valuation ranges.

What alternative data sources reveal private company health?

The signal-mining agent examines hiring trends on LinkedIn and Boss Zhipin, app store rankings and reviews, patent filings, litigation records, business registration changes, and secondary share trading. These signals are cross-checked against the company's public narrative for inconsistencies.

Does this research framework work for companies outside China?

Yes, it searches both Chinese and English sources including SEC filings, Bloomberg, The Information, and TechCrunch. It was designed for companies like SpaceX and Stripe as well as Chinese firms such as Ant Group and Xiaohongshu.

What are the limitations of AI-based private company research?

Information scarcity can cause false precision or false conservatism, so the framework mandates explicit unknown labeling instead of filling gaps with speculation. When data is severely insufficient, it switches to first-principles mode and states that a reliable valuation cannot be given.