An AI agent skill designed for automated data cleaning, statistical testing, financial modeling, and scenario simulation following the CRISP-DM methodology.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install guanshi-data-expert
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install guanshi-data-expert using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is GuanShi Data Expert?
GuanShi Data Expert is a professional-grade AI agent skill built to handle data cleaning, statistical inference, financial modeling, and data visualization. Working as a core component of the GuanShi Expert Cluster, this skill transforms raw structured and semi-structured datasets into highly quantitative insights to back strategic decision-making. By leveraging standard Python scientific libraries, it automates end-to-end data preparation and analytical workflows.
Integrating this skill into your AI workflows via Openclaw Skills provides a reliable, reproducible pipeline for calculating NPV, conducting sensitivity analyses, performing Mann-Kendall trend tests, and generating publication-ready charts. Whether analyzing competitor revenues, evaluating regional market entries, or modeling scenario simulations, it acts as an embedded data scientist for your agent swarm.
GuanShi Data Expert Use Cases
- Automated Financial Feasibility Studies: Calculate NPV, IRR, and payback periods with built-in sensitivity analysis (varying key variables by ±20%) for market entry or project evaluation.
- Trend & Correlation Testing: Apply Mann-Kendall non-parametric tests for temporal trends and Pearson/Spearman matrices for identifying statistical relationships in market datasets.
- Multi-Scenario Modeling: Simulating optimistic, baseline, and pessimistic outlooks with assigned probability distributions and custom boundaries to manage business risks.
- Scientific Data Visualization: Generating clean, high-resolution (150+ DPI) charts including waterfall charts, scatter plots with trendlines, and grouped bar graphs using matplotlib.
How GuanShi Data Expert Works
- Data Ingestion: The skill reads raw files (CSV, Excel, or JSON) from the Workspace via the integrated file agent.
- CRISP-DM Cleaning: Runs automated quality checks, identifying outlier ranges using IQR, handling missing values dynamically (mean, forward fill, mode), and normalizing column naming.
- Statistical Inference & Modeling: Executes custom Python scripts utilizing numpy, pandas, and scipy to compute financial metrics or run hypothesis tests (t-test, ANOVA, Mann-Whitney U).
- Visualization Generation: Renders clean, Unicode-compatible visualizations, saving them with high density (>=150 DPI) in the designated output directory.
- Automated MD Reporting: Compiles results into a standardized Markdown report containing clean tables, embedded chart paths, statistical significance markers, and explicit limitation statements.
GuanShi Data Expert Setup
To get started with this component from the Openclaw Skills registry, ensure your Python environment is active and run the following command to make sure dependencies are satisfied:
pip install pandas numpy scipy matplotlib
The agent is configured automatically via its skill definition. Ensure the Python executor has path permissions to write inside the output and temp workspaces.
GuanShi Data Expert Data Schema & Taxonomy
The GuanShi Data Expert structures files and outputs using a strict taxonomy. It strictly enforces a read-only policy for original datasets.
Output File Naming Taxonomy
| File Type | Naming Template | Example |
|---|---|---|
| Charts & Visuals | {Topic}_{ChartType}_{YYYYMMDD_HHMM}.png |
smart_lock_market_trend_line_20260606_1430.png |
| Markdown Reports | {Topic}_data_analysis_report_{YYYYMMDD_HHMM}.md |
southeast_asia_entry_data_analysis_report_20260606_1430.md |
Temporary Data Schema
- Input Data: Read-only access to CSV, XLSX, and JSON files in workspace.
- Intermediate Data: Stored inside the
./tempdirectory during cleaning and computation stages. - Final Outputs: Rendered charts (max 500KB per image, DPI >= 150) and final markdown files are published to the
./outputworkspace.
GuanShi Data Expert Advanced Features
- Robust Statistical Assumption Checks: Automatically validates normality assumptions using Shapiro-Wilk (n <= 5000) or Anderson-Darling (n > 5000) tests prior to applying parametric statistics.
- Multi-Agent Orchestration: Seamlessly interfaces within the GuanShi Cluster, receiving datasets from the Intelligence Expert and passing quantitative outcomes to the Strategy and Financial Strategy Experts.
- Dynamic Fallback Execution: Features fallback logic allowing the LLM to execute manual arithmetic operations if the Python environment becomes temporarily unavailable, labeling outputs with precision warnings.
- Outlier Preservation Principle: Flags IQR outliers as anomalies but preserves them to protect structural market signals, generating side-by-side comparative reports with and without anomalies.
SKILL.md
Loading
Related Openclaw Skills
Douyin Data Query Method (douyin-data-method)
A standardized data query framework to retrieve and analyze Douyin video, user, search, trends, and live data using the MaxHub API.
Financial Audit Domain
A specialized financial risk analysis and audit reference library designed for A-share non-financial listed companies, integrating 50 indicators and 5 fraud red lines.
Medical Imaging and Interventional Diagnosis (9th Edition)
A comprehensive medical imaging and interventional radiology clinical decision-support and educational skill package for AI agents, containing 136 core competencies based on the authoritative 9th Edition Medical Imaging textbook.
Systematic Anatomy 10th Edition Skill Pack
An advanced Openclaw skill pack mapping 112 core anatomical proficiencies based on the authoritative People's Medical Publishing House Systematic Anatomy 10th Edition.
GuanShi Intelligence Expert
An automated AI intelligence agent designed to search, filter, extract, and cross-validate industry data to produce structured, high-confidence intelligence reports.
Video Transcript Extraction Method
An intelligent, platform-independent skill that extracts speech content from online videos and converts it into structured, timestamped markdown transcripts and summaries.
Featured*
