Talib Technical Analysis

通过 Python 调用 150+ TA-Lib 技术分析指标（均线、MACD、RSI、布林带等），支持多市场金融数据的技术面量化计算。

This skill is an instruction-only blueprint for computing TA-Lib indicators and building ZVT pipelines; it does not request secrets or install code. Before using it: 1) verify you have a dedicated Python 3.12+ virtual environment and do not run it in an environment containing sensitive credentials; 2) review seed.yaml and references/* files (they are included and instruct the agent to reload seed.yaml) to ensure the semantic locks and preconditions match your trading/operational policies; 3) expect the agent to run small python checks (import zvt, test kdata access) and to touch a writable ZVT_HOME path (~/.zvt by default) — run those checks manually if you prefer; 4) if you plan to compute indicators using the TA-Lib C layer, audit your runtime for the anti-patterns listed (FFI dtype, initialization, NaN handling, thread safety) to avoid silent corruption; and 5) run any trading/backtest code in an isolated environment (sandbox or ephemeral VM) and do not provide unrelated cloud or broker credentials to this skill unless explicitly required and vetted.

Type: OpenClaw Skill Name: talib-technical-analysis Version: 0.3.3 The skill bundle is a highly structured framework for performing technical analysis using TA-Lib and the ZVT quantitative trading library. It contains extensive documentation, 'semantic locks,' and 'fatal constraints' (e.g., in SKILL.md and references/seed.yaml) designed to prevent common financial programming errors such as look-ahead bias, memory corruption in C-FFI layers, and incorrect data alignment. While the instructions for the AI agent are complex and prescriptive, they are entirely aligned with the stated purpose of ensuring the reliability and accuracy of financial indicators. No evidence of data exfiltration, malicious code execution, or harmful prompt injection was found.