Skill UI

Automates Luigi task generation for data pipelines, offering guided best practices, code and configuration outputs, and validation to streamline ETL, transformation, orchestration, and streaming workflows.

Automates ROC curve plotting tasks in ML training, guiding data prep, model training, tuning, tracking, and code generation for production-ready results.

High-performance Rust toolkit with Python bindings for genomic interval analysis, covering overlap detection, coverage track generation, ML tokenization, reference sequence handling, fragment processing, and scoring workflows for sequencing datasets.

Guides structured hypothesis generation by combining literature searches, evidence synthesis, competing explanations, and experimental design, with mandatory AI-generated scientific schematics for complete reports.

TorchDrug is a PyTorch-native toolbox for building graph neural networks tailored to molecules, proteins, and biomedical knowledge graphs, covering property prediction, generation, retrosynthesis, and knowledge graph reasoning with curated datasets and models.

This skill analyzes DNS query logs to detect sophisticated threats such as data exfiltration via DNS tunneling, Domain Generation Algorithm (DGA) usage, and covert Command and Control (C2) channels. It utilizes advanced techniques like entropy analysis, query volume anomaly detection, and subdomain length analysis within SIEM platforms, helping SOC teams identify threats that bypass traditional network security controls.

Automates the generation of high-quality, publication-ready figures and comparison tables from raw experimental data (JSON, CSV). This skill is essential for academic writing workflows, handling various plot types including line graphs, bar charts, scatter plots, and heatmaps, ensuring consistency with professional style guides like LaTeX and specific DPI requirements.

This skill provides a comprehensive workflow to transform raw data into professional visual charts. It intelligently analyzes data features (e.g., trends, comparisons, flows) to select the optimal chart type (line, bar, pie, etc.). Users provide data and specifications, and the skill generates a high-quality, detailed chart image using a structured JavaScript process.

Guides agents through validation, dry-run MCP tool usage, and monitors-as-code YAML generation so Monte Carlo monitors can be established via CLI or CI/CD without directly creating assets.

A specialized toolkit designed for generating and validating research-only artifacts for Clinical Decision Support (CDS) systems. It supports evidence-profile generation, cohort analysis, survival modeling, biomarker evaluation, and governance documentation. IMPORTANT: This skill is strictly for aggregate research and documentation; it must never be used for patient care, diagnosis, or live clinical operations.

Datamol is a high-level, Pythonic abstraction layer built over RDKit, designed for computational chemistry and drug discovery. It simplifies complex molecular operations, offering sensible defaults for tasks like SMILES standardization, descriptor calculation, fingerprint generation (ECFP4), 3D conformer analysis, and molecular clustering. It supports efficient parallel processing and cloud storage I/O, making it ideal for handling large-scale chemical datasets.

The HypoGeniC skill facilitates the iterative proposal and scoring of textual hypotheses from labeled datasets using LLMs. It provides tools for generating candidate hypothesis banks and task-prediction statistics. Crucially, it mandates a rigorous, local review workflow, emphasizing dataset auditing (checksums, schemas), configuration validation, and explicitly separating candidate results from definitive scientific conclusions.