Microsoft Qlib

Microsoft Qlib - AI-oriented Quantitative Investment Platform. Use when: (1) stock/financial data analysis, (2) quantitative trading strategy development, (3...

Security Scan

The skill's instructions and dependencies match the stated purpose (install and use Microsoft Qlib) but the package's provenance and a small metadata inconsistency are unclear and merit caution before install.

Purpose & Capability

The skill is coherent with a Qlib integration: it only requires Python and shows typical qlib usage (pip install qlib, qlib.init(), data fetching, backtesting). However, the registry lists the skill source as 'unknown' and provides no homepage while the SKILL.md claims Microsoft Qlib — this provenance mismatch could indicate impersonation or an untrusted publisher.

Instruction Scope

SKILL.md instructions are scoped to installing qlib and using its API (init, data access, backtest, model training). It does not instruct reading unrelated files, harvesting env vars, or sending data to unexpected endpoints.

Install Mechanism

Installation is via pip install qlib (declared in SKILL.md metadata). Pip installation from PyPI is expected for this package but carries the usual risk of executing third-party code. Also note a minor inconsistency: registry metadata said 'No install spec' while SKILL.md contains an openclaw.install entry — this mismatch should be reconciled.

Credentials

No environment variables or credentials are requested by the skill. This is proportionate for an instruction-only qlib integration. Be aware that some data sources used with qlib may require separate API keys which are not requested here.

Persistence & Privilege

The skill is not force-enabled (always:false) and is user-invocable. It does not request elevated platform privileges or persistent presence.

What to consider before installing

This SKILL appears to be a plain Qlib usage guide and pip-based install, which is coherent with its description — but two things to check before installing: (1) provenance: the registry entry has no source/homepage while the skill claims 'Microsoft Qlib' — confirm the publisher is trustworthy and that you are installing the official qlib package (compare with the official GitHub and PyPI entries). (2) pip risk: pip install will download and run code from PyPI; install into an isolated virtualenv or container and review the package/version. Finally, be aware that real data access may require separate data-provider API keys or configurations handled outside this skill.

Microsoft Qlib Skill

Microsoft Qlib is an AI-oriented quantitative investment platform developed by Microsoft Research.

Features

• Data Handling: Professional financial data processing and management
• Alpha Factors: Advanced factor mining and analysis
• Backtesting: High-performance backtesting engine
• Portfolio Optimization: Risk modeling and portfolio optimization
• Machine Learning: ML models for financial predictions

When to Use

This skill is perfect for:

1. Stock/Financial Data Analysis - Analyze historical stock data, financial indicators
2. Quantitative Trading Strategy Development - Build and test trading algorithms
3. Backtesting Trading Strategies - Validate strategies with historical data
4. Machine Learning for Finance - Train ML models for price prediction
5. Portfolio Optimization - Optimize portfolio allocation and risk management
6. Risk Modeling - Assess and model financial risks
7. Fetching Stock Prices - Retrieve real-time and historical stock data
8. Factor Analysis - Alpha factor mining and research
9. Model Training - Train predictive models for financial markets

Installation

pip install qlib

Quick Start

Initialize Qlib

import qlib
qlib.init()

Fetch Stock Data

from qlib.data import D

# Get stock features
df = D.features(
    instruments=["AAPL", "MSFT"],
    fields=["Close($close)", "Volume($volume)"],
    freq="day"
)

Build Strategy

from qlib.workflow import R
from qlib.contrib.evaluate import backtest_daily

# Create and run strategy
with R.start(experiment_name="my_strategy"):
    # Strategy implementation
    result = backtest_daily(start_time="2020-01-01", end_time="2023-12-31")

Train Model

from qlib.contrib.model.gbdt import LGBModel

# Initialize model
model = LGBModel()
model.fit(dataset_train)
pred = model.predict(dataset_test)

Requirements

• Python 3.7+
• pip

Resources

• GitHub: https://github.com/microsoft/qlib
• Documentation: https://qlib.readthedocs.io/
• Official Site: https://microsoft.github.io/qlib/

Notes

• This skill requires Python environment with pip installed
• Qlib is maintained by Microsoft Research
• For best performance, use with sufficient memory (recommended 8GB+)