finance-bottleneck-hunter

Scans global supply chains for physical bottlenecks and screens listed companies for arbitrage opportunities.

Updated Aug 10, 2026
One-click install
npx skills add https://github.com/Choi-Keith/skill-arsenal-ultra --skill finance-bottleneck-hunter-choi-keith
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: finance-bottleneck-hunter
Source: https://github.com/Choi-Keith/skill-arsenal-ultra/tree/main/plugins/finance-skills/finance-screening/skills/finance-bottleneck-hunter
Command: npx skills add https://github.com/Choi-Keith/skill-arsenal-ultra --skill finance-bottleneck-hunter-choi-keith

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Investors chasing super-trends like AI infrastructure often pile into already-priced leaders (GPU, HBM) while missing the real alpha in second- and third-layer supply chain chokepoints. This Skill systematically decomposes a super-trend into physical supply chain layers, identifies bottleneck segments with concentrated supply and long expansion cycles, and screens listed companies sitting on those chokepoints with mandatory valuation checks. ## Core Features & Use Cases - Supply Chain Decomposition: Breaks a super-trend (AI infrastructure, energy transition, defense, semiconductors, space) into Layer 0-4 physical components, focusing on under-covered Layer 2-3 segments like optical modules, InP substrates, ABF substrates, and probe cards. - Bottleneck Scoring & Company Screening: Rates each segment on 6 criteria (supplier concentration, expansion cycle, substitutability, utilization, demand growth, qualification cycle) to produce S/A/B bottleneck ratings, then screens listed companies by purity, market cap, and liquidity. - Mandatory Valuation Gates: Enforces red/yellow/green valuation checks (PS, PE, TAM ratio, 10-year exit return test) so a real bottleneck at 100x PS is flagged as overpriced rather than recommended. - Taiwan Stock Data Integration: Uses the bundled FinMind-based script to pull quotes, valuation, financials, and monthly revenue for Taiwan-listed suppliers, where monthly revenue YoY is the fastest public signal of bottleneck pricing power. - Use Case: Ask it to scan the AI infrastructure supply chain; it produces a bottleneck map, a ranked opportunity board with PS/PE and signal-strength ratings, one-page company summaries, and an hourly scan mode that only writes reports when new signals appear. ## Quick Start Ask the AI to run a supply chain bottleneck scan on the AI infrastructure trend and output a ranked list of bottleneck companies with valuation checks.

Frequently Asked Questions about finance-bottleneck-hunter

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

FAQPage Schema
How do I find supply chain bottleneck stocks in the AI infrastructure trend?▼

Decompose the trend into physical layers (components, materials, equipment, infrastructure), then score each Layer 2-3 segment on supplier concentration, expansion cycle, and demand growth. Segments with few suppliers and long capacity lead times are bottlenecks; screen listed companies there for revenue purity and valuation.

What criteria define a supply chain bottleneck for investment screening?▼

Six criteria: three or fewer global suppliers, capacity expansion over two years, no viable substitutes, utilization above 90%, demand growth above 50% annually, and customer qualification cycles over one year. Four or more red flags rate an S-level bottleneck.

How do I get Taiwan stock financial data and monthly revenue in Python?▼

Use the bundled twstock_data.py script, which queries the FinMind API with zero external dependencies. Commands cover quotes, valuation (PER/PBR), five-year financials, monthly revenue with YoY, dividends, and stock search by name or code.

Does the FinMind API require an API token for Taiwan stock data?▼

No, anonymous access works with hourly rate limits. To raise limits, set the FINMIND_TOKEN environment variable or place a token in a local file; invalid tokens return a clear error and can be removed to fall back to anonymous access.

Why is a real supply chain bottleneck not always a good investment?▼

A genuine bottleneck can already be fully priced in. The workflow enforces valuation gates: market cap above 20% of TAM, PS above 30x without 100% growth, or a 10-year exit return below 10% all cap the signal rating regardless of bottleneck purity.

When should I not use bottleneck scanning for stock research?▼

It is not suited for deep single-company fundamental research or traditional valuation analysis of established firms. It targets thematic, supply-constrained opportunities in physical hardware chains, not software upgrades or narrative-driven concepts.