Quantaxis Data Platform

提供 A 股市场的因子计算、存储与 tear sheet 分析能力，支持 Pandas/Polars 零拷贝数据转换和 QIFI 账户回测模拟，适用于多数据源量化研究。

This skill is instruction-only and appears coherent for A‑share/factor research and backtesting. Before installing or letting the agent execute commands: (1) Review any proposed shell commands (especially pip install) and prefer using a virtualenv/container to avoid global package changes. (2) Expect the agent to ask for data-source credentials (eastmoney/joinquant/qmt) or DB connection strings later — do not share secrets until you confirm the exact use and endpoint. (3) Be prepared for disk writes (ZVT_HOME ~/.zvt) and potentially large data pulls (ClickHouse, full‑market downloads) — run initial experiments on a small universe. (4) If you need stricter isolation, run the skill actions in a disposable container or VM and inspect commands the agent requests before consenting.

Type: OpenClaw Skill Name: quantaxis-data-platform Version: 0.3.3 The quantaxis-data-platform bundle is a legitimate quantitative trading research tool focused on the A-share market. It employs sophisticated prompt-engineering techniques, such as 'Semantic Locks' (SL-01 to SL-12) and 'Fatal Constraints' (finance-C-*), to ensure the AI agent generates realistic trading code that respects financial regulations like T+1 settlement and avoids common errors like look-ahead bias. The bundle includes a valuable database of cross-project anti-patterns (e.g., AP-QLIB-1930, AP-ZVT-183) to protect users from known bugs in popular quant libraries. No evidence of data exfiltration, malicious execution, or harmful instructions was found; the complex logic is entirely dedicated to enforcing domain-specific correctness and safety in financial simulations.