Description

A 股量化实验室：基于 zvt 框架的数据采集 + 因子研究 + 回测执行一站式。覆盖 31 个场景——机构持仓、财报、指数成分、MACD/MA/量能择时。仅限中国 A 股。

README (SKILL.md)

A 股量化实验室 (a-stock-quant-lab)

Name: A Stock Quant Lab
Author: tangweigang-jpg

说出"跟机构持仓"或"MACD 回测"——我基于 zvt 直接写代码跑起来，不用你翻文档。美股数据质量一般，不推荐。

Pipeline

data_collection -> data_storage -> factor_computation -> target_selection -> trading_execution -> visualization

Top Use Cases (31 total)

Actor Data Recorder (`UC-101`)

Collects institutional investor holdings and top 10 free float shareholders on a weekly schedule for tracking major player positions Triggers: institutional investor, top holders, actor data

Financial Statement Recorder (`UC-102`)

Collects fundamental financial data including balance sheets, income statements, and cash flow statements from eastmoney on a weekly basis Triggers: financial statements, balance sheet, income statement

Index Data Recorder (`UC-103`)

Collects index metadata, index compositions (SZ1000, SZ2000, growth, value indices), and daily index price data Triggers: index data, index composition, SZ1000

For all 31 use cases, see references/USE_CASES.md.

Execute trigger: When user intent matches intent_router.uc_entries[].positive_terms AND user uses action verb (run/execute/跑/执行/backtest/fetch/collect)

What I'll Ask You

Target market: A-share (default), HK, or crypto? (US stocks in ZVT are half-baked — stockus_nasdaq_AAPL exists but coverage is thin)
Data source / provider: eastmoney (free, no account), joinquant (account+paid), baostock (free, good history), akshare, or qmt (broker)?
Strategy type: MACD golden-cross, MA crossover, volume breakout, fundamental screen, or custom factor?
Time range: start_timestamp and end_timestamp for backtest period
Target entity IDs: specific stocks (stock_sh_600000) or index components (SZ1000)?

Semantic Locks (Fatal)

ID	Rule	On Violation
`SL-01`	Execute sell orders before buy orders in every trading cycle	halt
`SL-02`	Trading signals MUST use next-bar execution (no look-ahead)	halt
`SL-03`	Entity IDs MUST follow format entity_type_exchange_code	halt
`SL-04`	DataFrame index MUST be MultiIndex (entity_id, timestamp)	halt
`SL-05`	TradingSignal MUST have EXACTLY ONE of: position_pct, order_money, order_amount	halt
`SL-06`	filter_result column semantics: True=BUY, False=SELL, None/NaN=NO ACTION	halt
`SL-07`	Transformer MUST run BEFORE Accumulator in factor pipeline	halt
`SL-08`	MACD parameters locked: fast=12, slow=26, signal=9	halt

Full lock definitions: references/LOCKS.md

Top Anti-Patterns (47 total)

AP-ZVT-183: 除权因子为 inf/NaN 时直接参与乘法导致复权静默失败
AP-ZVT-179: 第三方数据接口超限后异常被吞噬，数据静默缺失
AP-ZVT-200: Token 失效后数据查询返回空 DataFrame 而非报错

All 47 anti-patterns: references/ANTI_PATTERNS.md

Evidence Quality Notice

[QUALITY NOTICE] This crystal was compiled from blueprint finance-bp-009. Evidence verify ratio = 55.0% and audit fail total = 36. Generated results may have uncaptured requirement gaps. Verify critical decisions against source files (LATEST.yaml / LATEST.jsonl).

Reference Files

File	Contents	When to Load
references/seed.yaml	V6+ 全量权威 (source-of-truth)	有行为/决策争议时必读
references/ANTI_PATTERNS.md	47 条跨项目反模式	开始实现前
references/WISDOM.md	跨项目精华借鉴	架构决策时
references/CONSTRAINTS.md	domain + fatal 约束	规则冲突时
references/USE_CASES.md	全量 KUC-* 业务场景	需要完整示例时
references/LOCKS.md	SL-* + preconditions + hints	生成回测/交易代码前
references/COMPONENTS.md	AST 组件地图（按 module 拆分）	查 API 时

Compiled by Doramagic crystal-compilation-v6.1 from finance-bp-009 blueprint at 2026-04-20T07:34:47.524525+00:00. See human_summary.md for non-technical overview.

Usage Guidance

This skill is largely what it says: an instruction-only A‑share quant lab built around zvt that will run Python commands, download data from external providers, and write to a local zvt data directory (~/.zvt). Before installing: (1) Verify the source — this skill has no homepage and an unknown owner. (2) Confirm whether you have or are willing to provide any required API tokens (joinquant, broker/qmt) and where they should be stored — the skill does not declare these env vars. (3) Investigate the declared required binary 'uv' (not typical for zvt) — ask the publisher what 'uv' is and why it's required. (4) Expect the skill to run pip installs and create/write files under ZVT_HOME (~/.zvt) — if you want to limit its reach run it in an isolated environment/virtualenv or sandbox. (5) If you will give the agent autonomous invocation rights and network access, be aware it could perform data downloads and write files without additional prompts; reduce privileges if you are uncomfortable. If the publisher can clarify the 'uv' requirement and supply a list of exact env vars/credentials and config paths the skill will use, that will resolve the main inconsistencies.

Capability Analysis

Type: OpenClaw Skill Name: a-stock-quant-lab Version: 0.1.2 The skill bundle is a comprehensive instruction set and documentation package for a quantitative trading agent specializing in the Chinese A-share market using the zvt framework. It includes detailed domain constraints (e.g., T+1 settlement, price limits), technical anti-patterns for various quant libraries (qlib, vnpy, zipline), and 'Semantic Locks' designed to prevent the AI from generating logically flawed or biased backtesting code (e.g., look-ahead bias). No evidence of data exfiltration, malicious execution, or harmful prompt injection was found; the content is strictly aligned with financial research and data sourcing.

Capability Tags

cryptorequires-oauth-tokenrequires-sensitive-credentials

Capability Assessment

ℹ Purpose & Capability

Name and description match the instructions: this is an instruction-only wrapper around zvt for A‑share data collection, factor research and backtesting. Asking for python3 is coherent. However the declared required binary 'uv' is unusual for a Python-only zvt workflow and is not explained in SKILL.md, which is disproportionate/unexpected.

⚠ Instruction Scope

SKILL.md instructs the agent to run Python commands (pip install zvt, zvt.recorders, zvt.init_dirs), read/write the ZVT home (~/.zvt) and perform network calls to multiple external data providers (eastmoney, joinquant, baostock, akshare). The skill metadata claims no required config paths or env vars, yet runtime preconditions explicitly reference ZVT_HOME and attempt to write to the user's home directory; the instructions therefore access filesystem state and environment variables not declared in the registry metadata.

ℹ Install Mechanism

No install spec (instruction-only), which is low risk by itself. The SKILL.md still expects the environment to allow pip installs (e.g., 'pip install zvt') and to run arbitrary Python code. There is no remote download or hidden installer, but the skill will instruct the agent to fetch packages and network data at runtime.

⚠ Credentials

The skill declares no required environment variables, yet it expects network access to third‑party providers; some providers (joinquant, qmt broker) commonly require API keys/accounts but no credentials or tokens are declared. Primary credential is listed as 'python' (not a secret) which is not a real credential. Missing declaration of expected secrets/config (e.g., JOINQUANT_TOKEN, QMT credentials, or ZVT_HOME path) is an incoherence.

ℹ Persistence & Privilege

always:false and model invocation allowed (platform default). The instructions will create and write to a local zvt home directory (~/.zvt) and may persist downloaded data; writing to the user's data directory is expected for this purpose but the registry metadata did not declare required config paths, which is a discrepancy to review.

Version History

v0.1.2

v0.1.2: inject bilingual metadata per naming spec. H1 now shows A 股量化实验室 + slug; tagline and description replaced with CTO-authored copy (fixes tagline pollution for non-ZVT skills).

v0.1.1

Remove install.sh — skill is now a knowledge-only bundle. Host AI consumes directly from URL; no user-side installation needed. Fixes ClawHub suspicious flag.

v0.1.0

初版发布。A 股量化实验室——基于 zvt 框架的数据采集、因子研究、回测全流水线。31 UCs / 8 semantic locks / 47 anti-patterns / 424-class zvt 组件地图。仅限 A 股，MVP 验证期。源码：https://github.com/tangweigang-jpg/doramagic-skills

Metadata

Slug a-stock-quant-lab

Version 0.1.2

License MIT-0

All-time Installs 0

Active Installs 0

Total Versions 3

Frequently Asked Questions

What is A Stock Quant Lab?

A 股量化实验室：基于 zvt 框架的数据采集 + 因子研究 + 回测执行一站式。覆盖 31 个场景——机构持仓、财报、指数成分、MACD/MA/量能择时。仅限中国 A 股。 It is an AI Agent Skill for Claude Code / OpenClaw, with 125 downloads so far.

How do I install A Stock Quant Lab?

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

Is A Stock Quant Lab free?

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

Which platforms does A Stock Quant Lab support?

A Stock Quant Lab is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created A Stock Quant Lab?

It is built and maintained by Tang Weigang (@tangweigang-jpg); the current version is v0.1.2.

More Skills

A Stock Quant Lab