Adr Hshare logo

Adr Hshare

OrganizationPopular
HKUDS
adr-hshare

ADR/H-share/A-share cross-listing premium analysis — track pricing gaps between US-listed ADRs, HK-listed H-shares, and A-shares for arbitrage signals, dual-listing valuation, and delisting risk assessment.

Overview

PublisherHKUDS
RepositoryVibe-Trading
Skill nameadr-hshare
Stars
33.6K
Forks
5.5K
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by HKUDS on GitHub. Read the source before you install it.

Installation

Install the Adr Hshare AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/HKUDS/Vibe-Trading.git /tmp/Vibe-Trading
mkdir -p .claude/skills
cp -r /tmp/Vibe-Trading/agent/src/skills/adr-hshare .claude/skills/adr-hshare
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Adr Hshare in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Adr Hshare on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Adr Hshare is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

ADR / H-Share / A-Share Cross-Listing Analysis

Overview

Many Chinese companies are listed on multiple exchanges — A-shares in Shanghai/Shenzhen, H-shares in Hong Kong, and ADRs in the US. Pricing gaps between these listings create arbitrage opportunities and reveal market-specific sentiment differences. This skill provides frameworks for analyzing cross-listing premiums, identifying arbitrage signals, and assessing delisting risk for US-listed Chinese ADRs.

Core Concepts

1. Cross-Listing Structures

StructureDescriptionExamples
A + H dual-listedSame company listed on both A-share and HK exchangePetroChina (601857.SH / 0857.HK), ICBC (601398.SH / 1398.HK)
H + ADR dual-listedHK-listed with US ADRAlibaba (9988.HK / BABA), JD.com (9618.HK / JD)
A + H + ADR triple-listedAll three marketsChina Life (601628.SH / 2628.HK / LFC)
HK primary + US secondaryPrimary listing in HK, secondary ADRTencent (0700.HK / TCEHY OTC)
US primary → HK secondaryOriginally US, added HK listingAlibaba (BABA → 9988.HK), Baidu (BIDU → 9888.HK)

2. AH Premium Analysis

AH Premium = (A-share price / H-share price in CNY terms - 1) × 100%

python
def calculate_ah_premium(a_price_cny, h_price_hkd, usdcny, usdhkd):
    """Calculate AH premium for a dual-listed stock."""
    h_price_cny = h_price_hkd * (usdcny / usdhkd)  # Convert HKD to CNY
    ah_premium = (a_price_cny / h_price_cny - 1) * 100
    return ah_premium

# Example: PetroChina
# A-share: 8.50 CNY, H-share: 6.20 HKD
# USDCNY: 7.25, USDHKD: 7.82
# H in CNY: 6.20 * (7.25/7.82) = 5.75 CNY
# AH Premium: (8.50/5.75 - 1) * 100 = 47.8%

AH Premium signal interpretation:

Premium LevelInterpretationAction
>50%Extreme A-share premium; A-share speculative bubble or H-share extreme undervaluationStrong: buy H, sell/avoid A
30-50%Elevated premium; normal for high-retail-participation namesModerate: favor H if fundamentals same
10-30%Normal range for most AH pairsNeutral; no strong arbitrage signal
0-10%Compressed premium; A-shares relatively cheapUnusual; investigate catalyst
<0%H-share premium over A-shareVery rare; usually near-term event-driven

Structural drivers of AH premium:

  1. Liquidity premium: A-shares have much higher retail participation and turnover → liquidity premium
  2. Access premium: A-shares were historically hard for foreigners to access → scarcity premium
  3. Currency expectations: CNY depreciation expectations widen the premium
  4. Regulatory arbitrage: different trading rules (T+1 in A-shares vs T+0 in HK)
  5. Investor composition: A-share retail speculative premium vs HK institutional valuation discipline

3. ADR Premium/Discount Analysis

ADR premium = (ADR price in USD / HK equivalent in USD - 1) × 100%

python
def calculate_adr_premium(adr_price_usd, hk_price_hkd, adr_ratio, usdhkd):
    """
    Calculate ADR premium over HK listing.
    adr_ratio: number of HK shares per 1 ADR (e.g., BABA: 1 ADR = 8 HK shares)
    """
    hk_equivalent_usd = (hk_price_hkd * adr_ratio) / usdhkd
    premium = (adr_price_usd / hk_equivalent_usd - 1) * 100
    return premium

# Example: Alibaba
# BABA ADR: $85.00, 9988.HK: HKD 82.50
# ADR ratio: 1 ADR = 8 HK shares
# HK equivalent: (82.50 * 8) / 7.82 = $84.40
# ADR premium: (85.00/84.40 - 1) * 100 = 0.71%

Key ADR conversion ratios:

CompanyADR TickerHK TickerADR Ratio (HK:ADR)ADR Exchange
AlibabaBABA9988.HK8:1NYSE
JD.comJD9618.HK2:1NASDAQ
BaiduBIDU9888.HK8:1NASDAQ
BilibiliBILI9626.HK1:1NASDAQ
NIONIO9866.HK1:1NYSE
XPengXPEV9868.HK2:1NYSE
Li AutoLI2015.HK2:1NASDAQ
NetEaseNTES9999.HK5:1NASDAQ
Trip.comTCOM9961.HK1:1NASDAQ
PinduoduoPDDN/A (US-only)N/ANASDAQ

ADR premium drivers:

  • US trading hours sentiment (earnings releases, macro data during US hours)
  • US-specific regulatory events (SEC, PCAOB audits)
  • Liquidity premium (ADR often more liquid for global funds)
  • Time zone gap: ADR closes at HK's open → overnight gap creates premium/discount

4. Delisting Risk Assessment

HFCAA (Holding Foreign Companies Accountable Act) framework:

Since 2022, PCAOB gained access to audit workpapers of Chinese companies. Key risks:

Risk LevelCriteriaImpact
LowPCAOB inspection completed, no issuesADR status stable
MediumPCAOB inspection completed, deficiencies notedMonitor for resolution
HighPCAOB access revoked or restricted3-year delisting countdown activated
CriticalOn SEC "identified issuer" list for 3 consecutive yearsForced delisting

Delisting risk indicators:

python
delisting_risk_factors = {
    "pcaob_status": "inspected",     # inspected / pending / blocked
    "sec_identified_years": 0,        # 0, 1, 2, or 3 (3 = delist)
    "has_hk_listing": True,           # Backup listing reduces impact
    "hk_listing_type": "primary",     # primary (can be in Connect) vs secondary
    "vie_structure": True,            # Variable Interest Entity adds legal risk
    "state_owned": False,             # SOE status adds geopolitical risk
}

# Companies with HK primary listing (BABA, JD, BIDU, NTES, etc.) have
# a safety net if US delisting occurs → fungible conversion ADR → HK shares
# Companies with US-only listing (PDD until HK listing) face higher risk

5. Cross-Listing Arbitrage Strategies

Strategy 1: AH Premium Mean-Reversion

python
# When AH premium for a specific stock diverges significantly from its historical average
ah_premium_current = 45  # current premium
ah_premium_mean_12m = 35  # 12-month average
ah_premium_std = 8        # standard deviation

z_score = (ah_premium_current - ah_premium_mean_12m) / ah_premium_std

if z_score > 2.0:
    signal = "fade_premium"  # A-share overvalued vs H; buy H, avoid A
elif z_score < -2.0:
    signal = "buy_premium"   # A-share undervalued vs H; buy A, avoid H
else:
    signal = "neutral"

Strategy 2: ADR-HK Intraday Arbitrage

  • During overlapping trading hours (HK morning = US pre-market via ADR), price gaps can appear
  • Professional arbitrageurs use ADR↔HK fungible conversion to capture these gaps
  • For research purposes: tracking ADR premium trend indicates which market is leading sentiment

Strategy 3: Event-Driven Cross-Listing

  • New HK listing announcement (US→HK): ADR typically dips 2-5% on dilution fear, then recovers
  • MSCI / FTSE index inclusion of HK listing: triggers passive fund buying in HK
  • Stock Connect inclusion (HK primary listing eligible): triggers mainland institutional buying

Data Access

python
import yfinance as yf

# Fetch ADR and HK prices simultaneously
baba_adr = yf.download("BABA", start="2025-01-01", end="2026-03-30", progress=False)
baba_hk = yf.download("9988.HK", start="2025-01-01", end="2026-03-30", progress=False)

# For A+H pairs
petrochina_a = yf.download("601857.SS", start="2025-01-01", end="2026-03-30", progress=False)
petrochina_h = yf.download("0857.HK", start="2025-01-01", end="2026-03-30", progress=False)

# FX rates for premium calculation
cny = yf.download("CNY=X", start="2025-01-01", end="2026-03-30", progress=False)  # USD/CNY
hkd = yf.download("HKD=X", start="2025-01-01", end="2026-03-30", progress=False)  # USD/HKD

# AH Premium Index (HSAHP)
# Not directly on yfinance; use Hang Seng website or Tushare

Output Format

## Cross-Listing Analysis — [Company Name]

### Listing Structure
- **A-share**: [code] @ [price CNY]
- **H-share**: [code] @ [price HKD]
- **ADR**: [ticker] @ [price USD] (ratio: X HK shares = 1 ADR)

### Premium/Discount
- **AH Premium**: [X%] (12m avg: X%, z-score: X.X)
- **ADR-HK Premium**: [X%] (5d avg: X%)
- **Direction**: [AH premium widening / narrowing / stable]

### Valuation Comparison
| Metric | A-share | H-share | ADR |
|--------|---------|---------|-----|
| PE (TTM) | XX.X | XX.X | XX.X |
| PB | X.X | X.X | X.X |
| Dividend yield | X.X% | X.X% | X.X% |

### Delisting Risk (ADR)
- **PCAOB status**: [inspected / pending]
- **SEC identified years**: [0/1/2/3]
- **HK backup**: [yes-primary / yes-secondary / no]
- **Risk level**: [low / medium / high / critical]

### Arbitrage Signal
- **AH premium z-score**: [X.X] → [fade premium / neutral / buy premium]
- **Best market to buy**: [A / H / ADR] — rationale
- **Catalyst**: [index inclusion, Connect eligibility, earnings]

### Investment Implication
- **Preferred listing**: [H-share / ADR / A-share] for new position
- **Risk**: [delisting, FX, regulatory, liquidity]

Notes

  • AH premium arbitrage is not freely executable: A-shares and H-shares are NOT fungible (no direct conversion), so true arbitrage requires separate capital pools
  • ADR↔HK conversion IS possible for most dual-listed names (via depositary bank), but takes 2-3 business days and involves fees
  • Currency risk (CNY, HKD, USD) is a major driver of cross-listing premiums; always hedge or account for FX when comparing
  • VIE (Variable Interest Entity) structure adds a layer of legal risk for many Chinese ADRs; this is a structural risk, not a trading signal
  • Stock Connect eligibility requirements mean not all HK-listed Chinese companies are accessible to mainland investors
  • This framework is for research purposes only and does not constitute investment advice

Frequently asked questions

What does the Adr Hshare AI skill do?

ADR/H-share/A-share cross-listing premium analysis — track pricing gaps between US-listed ADRs, HK-listed H-shares, and A-shares for arbitrage signals, dual-listing valuation, and delisting risk assessment.

Why use Adr Hshare on TypingMind?

Because you install it once and use it with any model. Adr Hshare is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Adr Hshare in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/HKUDS/Vibe-Trading/tree/main/agent/src/skills/adr-hshare. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Adr Hshare?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Adr Hshare?

As many as you like. As long as a model supports skills, you can use Adr Hshare with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Adr Hshare AI skill free?

Yes. It is published on GitHub by HKUDS under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇