← Back to Skills Marketplace
95
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install shenmeng-a-stock-predictor
Description
基于实时价格和技术指标,智能预测A股股票明日走势并生成详细分析报告与可视化图表供参考。
Usage Guidance
This skill appears to do what it says (generate A‑share technical predictions and charts) but contains several concerning implementation choices. Before installing or running: (1) be aware the skill will attempt to verify/charge users via https://skillpay.me and will call charge endpoints automatically — it may abort if payment fails; (2) the code contains a hardcoded billing API key (embedded secret) — treat this as insecure and potentially leaking your service account if reused; (3) the SKILL.md does not document the SKILLPAY_USER_ID environment variable used at runtime; (4) the predictor currently uses mocked data by default while a fetch_data function exists that would invoke a local kimi_finance module via subprocess and create temp files — confirm how real data is fetched and ensure the required kimi_finance package is trusted; (5) consider contacting the author for clarification, remove the hardcoded key, and test in an isolated environment (no sensitive accounts) before granting access or using it with real funds.
Capability Analysis
Type: OpenClaw Skill
Name: shenmeng-a-stock-predictor
Version: 1.0.0
The skill implements a custom monetization and billing layer in `payment.py` that makes external network requests to `https://skillpay.me` using a hardcoded API key, which is a high-risk practice for credential management and user privacy. Additionally, `stock_predictor.py` uses `subprocess.run` to execute the `kimi_finance` module and employs the insecure `tempfile.mktemp()` function for data handling. While these behaviors appear to support the stated functionality of a paid stock prediction service, the combination of external billing calls, hardcoded secrets, and sub-process execution warrants a suspicious classification.
Capability Assessment
Purpose & Capability
Name/description claim: real‑time A股 predictions via Kimi Finance. Code: predictor class, chart generator and a billing integration. Using Kimi Finance via subprocess is consistent with the stated data source, but the predictor actually uses a mocked _mock_predict in practice (real fetch_data is present but unused in predict). The presence of an integrated payment module is consistent with the SKILL.md pricing, but embedding a billing API key in code is disproportionate and risky.
Instruction Scope
SKILL.md describes only query/analysis UX and lists Kimi Finance and common Python libs. It does not document that the skill will: (1) attempt to verify/charge the user at runtime via an external billing endpoint and may exit if payment fails; (2) read SKILLPAY_USER_ID environment variable. The code also contains a fetch_data function that invokes subprocess to run the kimi_finance module and writes temp files, but predict() currently uses mock data instead — that mismatch is unexpected and grants the skill discretion to run subprocesses and create temp files without clear need.
Install Mechanism
No install spec (instruction-only install) and no downloads. All code is bundled with the skill. Runtime network calls (requests to skillpay.me) and subprocess invocation of a kimi_finance module are present but executed only at runtime; there is no high‑risk installer or external archive download in the manifest.
Credentials
Declared requirements: none. Actual code: uses os.environ.get('SKILLPAY_USER_ID') for billing and contains a hardcoded BILLING_API_KEY (sensitive secret) pointing to https://skillpay.me. That key in source is disproportionate and insecure (should not be embedded). The skill will contact an external billing service and can block execution if payment verification fails. No other credentials are required, but the undocumented env var and the embedded secret are notable red flags.
Persistence & Privilege
Skill is not always:true, does not request system‑wide changes, and does not modify other skills. It does perform network calls to an external billing API and may exit the process on unpaid use, but it does not ask for elevated system persistence.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install shenmeng-a-stock-predictor - After installation, invoke the skill by name or use
/shenmeng-a-stock-predictor - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
A股股票预测助手 v1.0.0 – 初始发布
- 支持A股实时行情查询与走势预测
- 自动计算多项技术指标(KDJ、RSI、MACD、MA、BOLL、CCI等)
- 基于技术分析,生成明日预测和概率提示
- 输出可视化走势图及技术指标图表
- 明确风险提示,预测仅供参考
- 支持多种股票代码格式,覆盖沪深北三市
Metadata
Frequently Asked Questions
What is A股股票预测助手?
基于实时价格和技术指标,智能预测A股股票明日走势并生成详细分析报告与可视化图表供参考。 It is an AI Agent Skill for Claude Code / OpenClaw, with 95 downloads so far.
How do I install A股股票预测助手?
Run "/install shenmeng-a-stock-predictor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is A股股票预测助手 free?
Yes, A股股票预测助手 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does A股股票预测助手 support?
A股股票预测助手 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created A股股票预测助手?
It is built and maintained by shenmeng (@shenmeng); the current version is v1.0.0.
More Skills