Data & Analytics

Data visualization, report generation, SQL queries, and spreadsheet automation. Transform your AI agent into a data-savvy analyst that turns raw data into actionable insights.

AutoML 自动化机器学习技能 | Automated Machine Learning Skill. 基于 PyCaret 进行低代码机器学习建模，支持分类、回归、聚类、异常检测、时间序列预测、自然语言处理和关联规则挖掘等任务。 未来将集成更多 AutoML 库（如 AutoGluon、FLAML 等）。 当用户需要快速构建机器学习模型、自动化模型选择、超参数调优、模型集成、特征工程或进行 AutoML 实验时使用此技能。 适用于数据科学家、公民数据科学家、机器学习工程师和希望快速原型开发的人员。 触发关键词：AutoML、机器学习自动化、PyCaret、分类模型、回归模型、聚类、异常检测、时间序列、文本分类、模型调优、模型比较、特征选择、统计检验、显著性检验、A/B测试。 Trigger keywords in English: AutoML, automated machine learning, PyCaret, classification, regression, clustering, anomaly detection, time series forecast...

Complete spreadsheet methodology — data cleanup, transformation, analysis, dashboards, automation, and reporting. Works with CSV, Excel, Google Sheets, or any tabular data. Use when the user needs to clean messy data, build reports, create dashboards, automate recurring spreadsheet tasks, or transform data between formats.

Use this skill for processing and analyzing large tabular datasets (billions of rows) that exceed available RAM. Vaex excels at out-of-core DataFrame operations, lazy evaluation, fast aggregations, efficient visualization of big data, and machine learning on large datasets. Apply when users need to work with large CSV/HDF5/Arrow/Parquet files, perform fast statistics on massive datasets, create visualizations of big data, or build ML pipelines that do not fit in memory.

Model interpretability and explainability using SHAP (SHapley Additive exPlanations). Use this skill when explaining machine learning model predictions, computing feature importance, generating SHAP plots (waterfall, beeswarm, bar, scatter, force, heatmap), debugging models, analyzing model bias or fairness, comparing models, or implementing explainable AI. Works with tree-based models (XGBoost, LightGBM, Random Forest), deep learning (TensorFlow, PyTorch), linear models, and any black-box mo...

Weights & Biases platform for ML experiment tracking, hyperparameter optimization, and artifact management. Log metrics, visualize training runs, run sweeps for hyperparameter tuning, and version datasets and models.

Forecast time series data using Google's TimesFM foundation model with zero-shot prediction. Use when: forecasting sales or demand, predicting server metrics, financial time series analysis, anomaly detection without training custom models.

Apache Superset is an open-source data exploration and visualization platform. Learn Docker deployment, database connections, chart creation, dashboard building, SQL Lab usage, and programmatic access via the REST API.

Assists with building, evaluating, and deploying machine learning models using scikit-learn. Use when performing data preprocessing, feature engineering, model selection, hyperparameter tuning, cross-validation, or building pipelines for classification, regression, and clustering tasks. Trigger words: sklearn, scikit-learn, machine learning, classification, regression, pipeline, cross-validation.

Generate professional reports with charts, tables, visualizations, and structured narratives. Covers data-driven PDF/HTML reports, weekly status reports, executive summaries, and recurring team updates. Use when a user asks to create a report, generate a data report, build a dashboard report, write a weekly report, create a status update, or produce a team progress report.

Build tools and dashboards for prediction markets — Polymarket, Manifold, Kalshi, and Metaculus. Use when tasks involve fetching prediction market data, building probability dashboards, analyzing market liquidity, creating trading bots for prediction markets, visualizing event probabilities, or tracking forecasting accuracy. Covers both API integration and market analysis.

Expert guidance for Pandera, the Python library for validating pandas and Polars DataFrames with expressive schemas. Helps developers define data contracts, validate data pipelines, and catch data quality issues before they corrupt downstream systems.

Expert guidance for MotherDuck, the serverless analytics platform built on DuckDB that combines local and cloud query execution. Helps developers run SQL analytics on cloud-hosted data, share datasets, and build hybrid local-cloud data pipelines using DuckDB's familiar interface.

Metabase is an open-source business intelligence tool for creating dashboards and visualizations. Learn Docker deployment, database connections, creating questions and dashboards, embedding analytics, and API usage.

Automate Canva from an agent — via the AI Connector MCP server (generate, find, edit, export designs in natural language) or the Connect REST API (OAuth 2.0 + PKCE, brand template autofill, asset uploads, exports, folders, comments). Use when: connecting an agent to Canva, batch-producing social posts or decks from a brand template and a spreadsheet, exporting designs to PDF/PNG/PPTX/MP4, uploading assets into a Canva account, building a Canva integration, or when the user mentions "Canva", "...

Audit Joomla extension URL flows for server-side request forgery and unsafe redirects — attacker-influenced URLs reaching an HTTP client or a browser Location header, including cloud-metadata and internal-service reachability and open redirects. Use when reviewing outbound requests or redirect targets in a Joomla extension, or when explicitly invoked for an SSRF audit.

Audit Joomla filter, user-state, and user-input values that reach SQL queries for unsafe validation, escaping, or parameterisation. Use for SQL injection reviews of a Joomla extension, or when explicitly invoked for SQL filtering analysis. For missing authentication/authorisation/CSRF checks see audit-authz; for privilege mismatches, backend-task leakage, or unsafe uploads see audit-controller-exposure.

Audit Joomla extension responses, logs, exports, errors, mail, and caches for injection into a downstream interpreter or disclosure of sensitive data. Covers response headers, CSV formula injection, log forging, error disclosure, mail header injection, and cache isolation. Use when reviewing non-HTML output sinks in a Joomla extension, or when explicitly invoked for this audit. Browser script execution belongs to audit-xss.

Audit Joomla record access paths for IDOR, BOLA, ownership bypass, and cross-user data leakage — whether every attacker-selectable identifier is constrained to what the current user may see or change. Use when reviewing per-record data isolation in a Joomla extension, or when explicitly invoked for an object-level access audit. For whether an entry point has any check see audit-authz; for whether it checks the right permission see audit-controller-exposure.

Audit attacker-influenced filesystem paths and archive entries in Joomla extensions for traversal, Zip Slip, symlink escape, and arbitrary read/write/delete. Use when reviewing file reads, downloads, writes, deletes, includes, or archive extraction in a Joomla extension, or when explicitly invoked for a filesystem audit. File type acceptance and initial upload storage belong to audit-controller-exposure.

Build statistical intuition from basic probability to advanced inference.

Solve optimization, statistics, signal processing, and linear algebra problems with SciPy recipes and ready-to-run code.

Analyze, transform, and clean DataFrames with efficient patterns for filtering, grouping, merging, and pivoting.

Write fast, memory-efficient numerical code with arrays, broadcasting, vectorization, and linear algebra.