Electrolyte and brine chemistry guidance for NeqSim. USE WHEN: modeling produced water, scale prediction, CO2/H2S in aqueous systems, MEG/DEG injection, hydrate inhibitor dosing, or any system with ions, salts, or electrolytes. Covers SystemElectrolyteCPAstatoil setup, ion components, scale risk assessment, and brine handling patterns.
NeqSim API patterns and code recipes. USE WHEN: writing Java or Python code that uses NeqSim for thermodynamic calculations, process simulation, or property retrieval. Covers EOS selection, fluid creation, flash calculations, property access, equipment patterns, and unit conventions.
Reaction engineering patterns for NeqSim. USE WHEN: modeling chemical reactors (equilibrium, kinetic, CSTR, PFR), setting up reaction systems, analyzing conversion/selectivity, or designing reactor networks. Covers GibbsReactor (equilibrium), PlugFlowReactor (kinetic PFR), StirredTankReactor (CSTR), KineticReaction setup, and CatalystBed configuration.
Dynamic simulation guidance for NeqSim. USE WHEN: running transient simulations, modeling startup/shutdown, tuning PID controllers, analyzing pressure/level dynamics, performing blowdown/depressurization, or setting up measurement devices and control loops. Covers runTransient, DynamicProcessHelper, controller tuning, and dynamic equipment configuration.
Work with ROR (Research-Information-System) for DICOM data exploration, workflow setup, and series selection in research PACS environments. This skill should be used when exploring medical imaging datasets, setting up processing workflows, querying DICOM series/studies by modality or metadata, or filtering patients/studies for batch processing.
Use when implementing collision detection, trace queries, physics simulation, or physical interactions in Unreal Engine. Triggers on: 'collision', 'trace', 'LineTrace', 'line trace', 'overlap', 'physics', 'hit result', 'sweep', 'collision channel', 'physics body', 'Chaos', 'raytrace', 'OnHit', 'OnBeginOverlap'. See related skills for component architecture and AI navigation.
Implements ray picking and collision detection. Use when selecting objects, detecting block interactions, or optimizing raycaster performance.
Find visually or functionally similar products from an image, name, or description.
American Fuzzy Lop Plus Plus mutation-based fuzz testing for finding crashes, hangs, and security vulnerabilities in binary programs.
Comprehensive Karma test runner skill for browser-based JavaScript unit testing with Jasmine, Mocha, or QUnit frameworks, real browser execution, coverage reporting, and CI/CD pipeline integration.
Deep code scan for AI security issues — prompt injection, PII in prompts, hardcoded keys, unguarded agents.
Kubernetes alert triage — dedup via YT search, deep control plane investigation, auto-escalation for recurring/flapping/control-plane alerts.
Research competitors, analyze market positioning, identify differentiators, and understand competitive landscape to inform product strategy and positioning. Use when researching competitors, analyzing features, identifying market gaps, understanding pricing strategies, evaluating competitive advantages, researching market trends, or informing product decisions.
Cheminformatics toolkit for molecular analysis and virtual screening. Use for SMILES/SDF parsing, molecular descriptor calculation (MW, LogP, TPSA), fingerprint generation (Morgan/ECFP, MACCS, RDKit), Tanimoto similarity search, substructure filtering with SMARTS, drug-likeness assessment (Lipinski Ro5), chemical reaction enumeration, 2D/3D coordinate generation, and compound library profiling. For simpler high-level API, use datamol. Use RDKit when you need fine-grained control over sanitization, custom fingerprints, SMARTS queries, or reaction SMARTS.
Guide for performing Structure-Activity Relationship (SAR) analysis using RDKit. Covers core scaffold identification via Maximum Common Substructure (MCS), R-group decomposition, molecular alignment for visualization, activity heatmap generation, and interpretive SAR text output. For general cheminformatics operations, see rdkit-cheminformatics. For bioactivity data retrieval, see chembl-database-bioactivity.
Rowan is a cloud-based computational chemistry platform providing quantum chemistry calculations via a Python SDK. Use it to run geometry optimization, conformer generation, torsional scans, and energy minimization with DFT or semiempirical methods, and retrieve molecular properties (dipole moment, partial charges, frontier orbital energies) — without managing local quantum chemistry software or HPC clusters.
Molecular featurization hub (100+ featurizers) for ML. Convert SMILES to numerical representations via fingerprints (ECFP, MACCS, MAP4), descriptors (RDKit 2D, Mordred), pretrained models (ChemBERTa, GIN, Graphormer), and pharmacophore features. Scikit-learn compatible transformers with parallelization, caching, and state persistence. For QSAR, virtual screening, similarity search, and deep learning on molecules.
Medicinal chemistry filters for drug discovery compound triage. Drug-likeness rules (Lipinski Ro5, Veber, Oprea, CNS, leadlike, REOS, Golden Triangle, Rule of Three), structural alerts (PAINS, NIBR, Lilly Demerits, Common Alerts), chemical group detection (hinge binders, Michael acceptors), molecular complexity metrics, property constraints, and a query language for composing filter logic. Built on RDKit and datamol. Use for hit-to-lead filtering, library design, and ADMET pre-screening. For molecular I/O and descriptors use rdkit-cheminformatics or datamol-cheminformatics instead.
Query PubChem database (110M+ compounds) via PubChemPy and PUG-REST API. Search compounds by name/CID/SMILES, retrieve molecular properties (MW, LogP, TPSA), perform similarity and substructure searches, access bioactivity data. For local cheminformatics computation use rdkit; for multi-database queries use bioservices.
Parse and query DrugBank local XML database for drug information, interactions, targets, and chemical properties. Search drugs by ID/name/CAS, extract drug-drug interactions with severity, map targets/enzymes/transporters, compute molecular similarity from SMILES. Primary access via local XML (downloaded); REST API available but rate-limited (3,000/month dev tier). For live bioactivity queries use chembl-database-bioactivity; for compound property lookups use pubchem-compound-search.
Therapeutics Data Commons (TDC) — AI-ready drug discovery dataset platform. Access curated ADME, toxicity, DTI, DDI datasets with scaffold/cold splits, standardized evaluation metrics, molecular oracles for optimization, and ADMET benchmark groups. Use for therapeutic ML model training, benchmarking, and molecular property prediction. For chemical database queries use chembl-database-bioactivity; for molecular featurization use molfeat.
Query the IUPHAR/BPS Guide to Pharmacology (GtoPdb) REST API for receptor-ligand interaction data, target pharmacology, and quantitative affinity metrics. Retrieve pKi/pIC50/pEC50 values, ligand classifications (approved drugs, biologics, natural products), target families (GPCRs, ion channels, nuclear receptors, kinases), and selectivity profiles across the pharmacological target space.
TorchDrug is a PyTorch-based machine learning platform for drug discovery. Use it for graph-based molecular representation learning, molecular property prediction (ADMET, activity), retrosynthesis prediction, drug-target interaction (DTI) modeling, and pretraining on large molecular datasets. Provides GNN layers (GraphConv, GAT, MPNN), pretrained models, and benchmark datasets in a unified PyTorch-compatible API.
