investment-team

Coordinates four parallel research agents to produce multi-perspective stock investment analysis reports.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Single-prompt AI stock analysis tends to produce vague, both-sides commentary without actionable conclusions. This Skill orchestrates a four-agent research team that analyzes a company from the perspectives of Buffett, Munger, Duan Yongping, and Li Lu, forcing explicit buy/wait/avoid conclusions with price ranges. ## Core Features & Use Cases - Parallel Multi-Agent Research: Launches four background agents covering business model, financials and valuation, industry competition, and risk and management quality, then synthesizes a final report. - Financial Rigor Verification: Requires exact arithmetic via tools/financial_rigor.py for market cap, valuation, cross-validation, and three-scenario valuation instead of LLM mental math. - Anti-Bias Safeguards: Includes information-richness grading (A/B/C), WebSearch permission pre-checks, dual-source data requirements, and a post-report audit sampling workflow via tools/report_audit.py. - Use Case: Ask for a deep-dive on a listed company such as Pinduoduo or Meituan and receive a structured report with four-dimension scoring, bull vs bear theses, a Buffett-style checklist, and tiered buy recommendations with price ranges. ## Quick Start Ask the assistant to run a full investment-team research analysis on a specific listed company and confirm the four-agent team structure to begin.

Frequently Asked Questions about investment-team

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

FAQPage Schema
How do I run a multi-agent stock analysis with this skill?▼

Provide a company name as the argument and confirm the four-agent team structure. The skill launches business, financial, industry, and risk analysts in parallel, then synthesizes their reports into a final investment recommendation with price ranges.

What investment frameworks does the four-agent analysis use?▼

Each agent applies a different value-investing lens: Duan Yongping for business model and moat, Buffett for financials and valuation, Munger for industry and competitive inversion thinking, and Li Lu for long-term certainty and management quality.

Why does the skill check WebSearch permissions before starting?▼

Background agents cannot prompt for interactive permission approval, so blocked WebSearch silently degrades them to training-knowledge-only answers. The pre-check verifies WebSearch is whitelisted in settings.local.json before launching any agent.

How does the skill verify financial data accuracy?▼

Financial data must come from two independent sources per market, and all valuation math runs through tools/financial_rigor.py for market cap, PE/PB verification, cross-validation, and three-scenario valuation. A final audit samples 15 percent of report figures via tools/report_audit.py.

What happens when public information about a company is scarce?▼

The skill assigns an information-richness grade of A, B, or C. For C-grade companies it switches to first-principles mode, focuses on core business questions, and explicitly labels data gaps rather than fabricating certainty.