Robo Advisor Python

自动化投资组合再平衡与交易执行，遵循先卖后买原则，支持多市场资产配置，智能计算最低交易规模及税费。。

This skill appears to be a legitimate robo-advisor blueprint, but there are important gaps you should address before installing or running it: 1) Confirm Python (3.12+) and the zvt package availability in your environment — the skill expects zvt and a writable ZVT_HOME (~/.zvt). 2) Ask the author which environment variables or API keys (joinquant/qmt/broker) are required and how they should be provided; do not paste secrets into chat. 3) Because the skill's instructions can run pip installs and Python commands and read workspace files, run it first in an isolated environment (container or VM) and perform dry-run/backtests only (no live orders) until you fully audit the behavior. 4) Review seed.yaml and the references/LOCKS.md locally to ensure the enforced 'semantic locks' (sell-before-buy, next-bar execution, T+1, MACD params) match your legal/regulatory and operational needs. 5) If you need higher assurance, request a version that explicitly declares required env vars, install steps, and a list of external endpoints so you can audit what will be contacted and what credentials are needed.

Type: OpenClaw Skill Name: robo-advisor-python Version: 0.3.3 The skill bundle is a highly structured framework for an AI agent to automate portfolio rebalancing and trading using the 'zvt' quantitative library. It contains extensive safety guardrails, including 'Semantic Locks' (e.g., SL-01, SL-02) and 'Fatal Constraints' (e.g., finance-C-001, finance-C-129) specifically designed to prevent financial logic errors like look-ahead bias, division by zero, and the use of hardcoded fallback credentials. The instructions in SKILL.md and seed.yaml are strictly aligned with the stated purpose of financial analytics and include robust output validation assertions (OV-01 to OV-06) to ensure the generated code remains within physically plausible and regulatory-compliant boundaries.