← Back to Skills Marketplace
chinasong

Hk Stock Predictor

by chinasong · GitHub ↗ · v1.0.0 · MIT-0
darwinlinuxwin32 ✓ Security Clean
337
Downloads
0
Stars
3
Active Installs
1
Versions
Install in OpenClaw
/install hk-stock-predictor
Description
Analyze Hong Kong listed stocks and produce prediction-ready theses from price action, fundamentals, technicals, southbound flows, AH premium, liquidity, and...
README (SKILL.md)

HK Stock Predictor

Research a Hong Kong listed stock or theme, then turn the analysis into clear, time-bounded predictions.

Minimal Input

User can provide any of:

{
  "symbol": "00700",
  "horizon": "30d"
}
{
  "theme": "南向资金持续流入的恒生科技成分股",
  "horizon": "14d"
}

Agent Normalization

Normalize input before analysis:

  • symbol: keep the 5-digit HK code when possible, also map to .HK form for external data.
  • companyName: resolve from ticker or user text.
  • horizon: normalize to one of 5d|14d|30d|90d|event.
  • benchmark: default to HSI or HSTECH for tech-heavy names.
  • predictionType: choose direction|range|event|relative-performance.

Preferred Data Stack

If these skills are available, use them in this order:

  1. akshare-skill for broad market and company data.
  2. hk-stock-analysis for HK-specific analysis framework.
  3. cross-border-flow-tracker for southbound flow and positioning.
  4. market-overview for index and sector context.

Fallback sources if the skills are not installed:

  • HKEX / HKEXnews
  • AAStocks
  • ET Net
  • AASTOCKS southbound / short selling pages
  • public financial data providers with clear source attribution

Deterministic Workflow

  1. Validate the target.
    • Reject symbols that cannot be resolved to a Hong Kong listed security.
    • If the user gives only a theme, narrow to 3-10 HK candidates first.
  2. Build market context.
    • Capture HSI, HSCEI, and HSTECH direction.
    • Note sector rotation, overnight macro drivers, and any major policy headline.
  3. Gather company facts.
    • Current price, market cap, 52-week range, valuation, earnings date, lot size, daily turnover.
  4. Gather HK-specific signals.
    • Southbound net buy/sell trend.
    • Short selling ratio and unusual changes.
    • AH premium or discount if dual-listed.
    • Liquidity warning if turnover is weak or spread is wide.
  5. Build the thesis.
    • State bull, base, and bear cases.
    • For each case, list 2-4 drivers and 1-2 invalidation signals.
  6. Convert thesis into forecastable statements.
    • Use binary, range, or relative-performance formats.
    • Every statement must be time-bounded and externally resolvable.
  7. Rank prediction candidates.
    • Prefer high observability, low ambiguity, and direct catalyst linkage.
    • Avoid questions that require subjective wording such as "表现好不好".
  8. Return the analysis and top prediction candidates.

Prediction Design Rules

Good prediction candidates:

  • "Will 00700 close above HK$520 on or before 2026-04-30?"
  • "Will 09988 outperform the Hang Seng Tech Index between now and 2026-05-15?"
  • "Will 02318 report YoY net profit growth above 10% in the next earnings release?"

Avoid:

  • vague targets without dates
  • subjective wording
  • multi-condition questions that resolve unclearly
  • tiny illiquid stocks where resolution may be distorted

Output

Return one structured object plus a readable summary.

{
  "ok": true,
  "normalizedInput": {
    "symbol": "00700",
    "symbolYahoo": "0700.HK",
    "companyName": "Tencent Holdings",
    "horizon": "30d",
    "benchmark": "HSTECH",
    "predictionType": "direction"
  },
  "marketContext": {
    "indices": [],
    "sectorTone": "",
    "macroDrivers": []
  },
  "evidence": {
    "fundamental": [],
    "technical": [],
    "flow": [],
    "hkSpecific": []
  },
  "scenarioAnalysis": {
    "bull": [],
    "base": [],
    "bear": []
  },
  "predictionCandidates": [
    {
      "title": "",
      "type": "direction|range|event|relative-performance",
      "deadlineIsoUtc": "",
      "resolutionSource": "",
      "confidence": 0,
      "why": []
    }
  ],
  "recommendedPrediction": {
    "title": "",
    "deadlineIsoUtc": "",
    "confidence": 0,
    "keyRisks": []
  },
  "warnings": []
}

Summary Template

# [公司名称] ([代码].HK) 港股推演

## 核心判断
- 方向判断:
- 时间窗口:
- 相对基准:

## 关键证据
- 基本面:
- 技术面:
- 资金面:
- 港股特有因子:

## 三种情景
- Bull:
- Base:
- Bear:

## 可预测题目
1. [候选题目 1]
2. [候选题目 2]
3. [候选题目 3]

## 首选题目
- 题目:
- 截止时间:
- 置信度:
- 主要风险:

If The User Wants Gougoubi Conversion

Convert the top prediction candidate into Gougoubi-ready fields:

  • marketName: use the selected prediction title.
  • deadlineIsoUtc: use the prediction deadline.
  • rules: include exact resolution source, timezone, comparison field, and tie handling.
  • tags: include hong-kong-stocks, sector tag, catalyst tag, and horizon tag.

Boundaries

  • Do not claim "all HK stocks" were checked unless the scan actually covered the full universe.
  • Do not hide missing data. Surface gaps in warnings.
  • Do not give investment advice phrased as certainty.
  • Prefer liquid names and observable events when generating prediction questions.
Usage Guidance
This package appears coherent for its stated purpose. Before installing: (1) Inspect any installer script referenced in INSTALL.md (do not run unknown shell scripts without review). (2) Verify the provenance — clawhub.json points to the HKEX site as 'repository', which is not a source code repo; confirm the skill author/host if you require provenance. (3) Be aware that using the skill may cause the agent to call other installed skills or fetch data from public sites (HKEX, AAStocks, etc.); those calls may require network access or provider credentials — only provide secrets to skills you trust. (4) After installation, run the skill in a sandboxed/test conversation first to confirm behavior and outputs. If you need stronger assurance about origin or integrity, request a published source repository or signed release before installing.
Capability Analysis
Type: OpenClaw Skill Name: hk-stock-predictor Version: 1.0.0 The hk-stock-predictor skill is a legitimate financial analysis tool designed to research Hong Kong stocks and generate structured prediction candidates. The instructions in SKILL.md focus on data normalization, market context gathering (HSI/HSTECH), and financial signal analysis (southbound flows, AH premium) using public data sources like HKEX and AAStocks. No evidence of malicious execution, data exfiltration, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
The name/description (HK stock prediction) match the SKILL.md workflow: ticker normalization, market context, southbound flows, AH premium, scenario-building and output templates. It does not request unrelated credentials or binaries.
Instruction Scope
SKILL.md stays on task and does not instruct reading arbitrary local files or exfiltrating secrets. It does recommend using companion skills (akshare-skill, hk-stock-analysis, cross-border-flow-tracker, market-overview) or public data sources (HKEX, AAStocks, ET Net). Be aware that using those companion skills or scraping listed public sites will involve network access and may in turn use credentials or different data providers; the SKILL.md itself does not direct data to unexpected external endpoints.
Install Mechanism
There is no automated install spec in the registry (instruction-only skill). The included INSTALL.md describes copying SKILL.md into ~/.codex/skills via a local script or a system installer script. The referenced scripts (scripts/install-hk-stock-predictor-skill.sh and the system install helper) are not part of the package manifest shown here — the install steps require running user-side scripts that write into the user's ~/.codex tree. This is reasonable but the user should inspect any installer script before running it.
Credentials
The package declares no required environment variables, no primary credential, and no sensitive config paths. That aligns with an instruction-only research skill. Note: if the agent calls companion skills, those other skills may require API keys or credentials (expected for data providers) — this skill itself does not request them.
Persistence & Privilege
always is false and the skill is user-invocable (normal). The only persistent action described is installing SKILL.md into ~/.codex/skills, which is a local file copy performed by the installer; the skill does not request elevated or system-wide privileges or modify other skills' configurations.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install hk-stock-predictor
  3. After installation, invoke the skill by name or use /hk-stock-predictor
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of hk-stock-predictor. - Analyzes Hong Kong-listed stocks using price action, fundamentals, technicals, southbound flows, AH premium, liquidity, and event catalysts. - Normalizes user input and builds structured, time-bounded, forecastable predictions. - Integrates with multiple data sources and prioritizes skills for HK-specific context. - Generates scenario analysis (bull/base/bear) and ranks prediction candidates. - Outputs include a structured object and a readable summary in both Chinese and English formats.
Metadata
Slug hk-stock-predictor
Version 1.0.0
License MIT-0
All-time Installs 3
Active Installs 3
Total Versions 1
Frequently Asked Questions

What is Hk Stock Predictor?

Analyze Hong Kong listed stocks and produce prediction-ready theses from price action, fundamentals, technicals, southbound flows, AH premium, liquidity, and... It is an AI Agent Skill for Claude Code / OpenClaw, with 337 downloads so far.

How do I install Hk Stock Predictor?

Run "/install hk-stock-predictor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Hk Stock Predictor free?

Yes, Hk Stock Predictor is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Hk Stock Predictor support?

Hk Stock Predictor is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux, win32).

Who created Hk Stock Predictor?

It is built and maintained by chinasong (@chinasong); the current version is v1.0.0.

💬 Comments