huang-sh

Code quality patterns and guidelines for ToolUniverse tool development. Apply when writing, fixing, or refactoring tool Python code in the ToolUniverse project. Encodes lessons from 80+ debug rounds. Use alongside devtu-fix-tool and devtu-self-evolve. Triggers: implementing tool fixes, writing new tool classes, reviewing tool code quality, checking schema correctness, looking up API-specific bug fixes.

The English summary is being prepared. Showing the original description.

Continuous improvement system for ToolUniverse tools, skills, and plugin. Run benchmarks, diagnose failures, route fixes to devtu skills, retest. Use after skill optimization, tool additions, or as regression check.

The English summary is being prepared. Showing the original description.

Strict SMILES validation, structural comparison, and modification verification. Catches invalid LLM-generated molecules.

The English summary is being prepared. Showing the original description.

Bayesian modeling with PyMC. Build hierarchical models, MCMC (NUTS), variational inference, LOO/WAIC comparison, posterior checks, for probabilistic programming and inference.

The English summary is being prepared. Showing the original description.

Multi-objective optimization framework. NSGA-II, NSGA-III, MOEA/D, Pareto fronts, constraint handling, benchmarks (ZDT, DTLZ), for engineering design and optimization problems.

The English summary is being prepared. Showing the original description.

Small molecule identification, characterization, and procurement — PubChem, ChEMBL, BindingDB, ADMET-AI, SwissADME, eMolecules, Enamine. Covers compound name to structure to activity to ADMET properties to commercial sourcing. Use for chemical biology, lead identification, probe selection, and the full small-molecule discovery pipeline.

The English summary is being prepared. Showing the original description.

Machine learning in Python with scikit-learn. Use when working with supervised learning (classification, regression), unsupervised learning (clustering, dimensionality reduction), model evaluation, hyperparameter tuning, preprocessing, or building ML pipelines. Provides comprehensive reference documentation for algorithms, preprocessing techniques, pipelines, and best practices.

The English summary is being prepared. Showing the original description.

Drug regulatory and approval research — FDA substance registry, ATC/EPC classification, EMA decisions, generic-drug status, FDA Orange Book exclusivity, NDA/BLA pathways. Use for jurisdiction-aware approval status (FDA vs EMA), generic vs brand availability, exclusivity expiry tracking, and regulatory pathway selection. Always specifies the market when reporting status.

The English summary is being prepared. Showing the original description.

Protein structure prediction from sequence. ESMFold-based, single GPU, no MSA needed. Predicts 3D structures with pLDDT confidence scores for drug discovery targets.

The English summary is being prepared. Showing the original description.

Enumerates virtual chemical libraries via reaction SMARTS transformations using RDKit and Reaction templates, with explicit handling of atom mapping, template extraction (RDKit reaction mining), product validation, RECAP/BRICS fragmentation, R-group decomposition, matched molecular pair analysis (MMPA), and Free-Wilson analysis. Use when generating combinatorial libraries from building blocks, enumerating analog series, deriving structure-activity rules, or extracting transformations from reaction data.

The English summary is being prepared. Showing the original description.

Complete mass spectrometry analysis platform. Use for proteomics workflows feature detection, peptide identification, protein quantification, and complex LC-MS/MS pipelines. Supports extensive file formats and algorithms. Best for proteomics, comprehensive MS data processing. For simple spectral comparison and metabolite ID use matchms.

The English summary is being prepared. Showing the original description.

Multi-method binding pocket detection and druggability assessment. Grid-based, fpocket, and P2Rank detection with druggability scoring, visualization, and cross-structure comparison.

The English summary is being prepared. Showing the original description.

Molecular featurization for ML (100+ featurizers). ECFP, MACCS, descriptors, pretrained models (ChemBERTa), convert SMILES to features, for QSAR and molecular ML.

The English summary is being prepared. Showing the original description.

Publication-quality molecular visualization. 2D structure drawings (PNG/SVG), molecule grids with property annotations, scaffold highlighting, protein-ligand interaction diagrams, and interactive 3D views.

The English summary is being prepared. Showing the original description.

Retrieve structurally similar compounds with known properties from ChEMBL/ZINC to ground predictions and inform optimization. Based on MolRAG (Xian 2025, ACL).

The English summary is being prepared. Showing the original description.

Iterative lead optimization with analyze-reason-generate-verify-evaluate loop. Paper-backed (MT-Mol, DrugR, MultiMol).

The English summary is being prepared. Showing the original description.

End-to-end molecular docking pipeline. Target preparation, pocket detection, protein-ligand docking (DiffDock/Vina), scoring, interaction analysis, and pose ranking.

The English summary is being prepared. Showing the original description.

Automated LLM-driven hypothesis generation and testing on tabular datasets. Use when you want to systematically explore hypotheses about patterns in empirical data (e.g., deception detection, content analysis). Combines literature insights with data-driven hypothesis testing. For manual hypothesis formulation use hypothesis-generation; for creative ideation use scientific-brainstorming.

The English summary is being prepared. Showing the original description.

Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.

The English summary is being prepared. Showing the original description.

Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters, use rdkit directly.

The English summary is being prepared. Showing the original description.

Performs structure-based virtual screening using AutoDock Vina, SMINA, GNINA (CNN scoring), and DiffDock-L hybrid workflows with explicit choice rules across rigid vs flexible docking, cross-docking vs self-docking, binding-site detection (P2Rank, fpocket), receptor preparation (PDB2PQR, PROPKA), ligand preparation (meeko, OpenBabel), and ultralarge-library screening (ZINC22, Enamine REAL). Use when screening chemical libraries against a protein target to find candidate binders, ranking docking poses, or selecting a docking workflow for a specific scenario.

The English summary is being prepared. Showing the original description.

Performs molecular similarity searching using Tanimoto, Tversky, Dice, and cosine coefficients on bit/count fingerprints with explicit choice rules for symmetric vs asymmetric measures, scaffold-hopping vs lead-optimization regimes, activity-cliff diagnosis, and large-library nearest-neighbor methods (BulkTanimoto, Annoy MHFP6, USRCAT). Use when ranking compounds by structural resemblance to a query, clustering libraries, finding analogs, or diagnosing activity cliffs.

The English summary is being prepared. Showing the original description.

Performs 3D shape-based similarity searching using ROCS (OpenEye), USRCAT (ultra-fast), Open3DAlign (RDKit), ESPSim (electrostatic), and ShaEP with explicit handling of Tanimoto-Combo (shape + color), shape vs ECFP4 complementarity, conformer-ensemble searching, alignment optimization, and scaffold hopping. Use when searching for shape-mimicking compounds with different scaffolds, identifying bioisosteric replacements, prospective scaffold hopping, or expanding hit series beyond 2D similarity.

The English summary is being prepared. Showing the original description.

Analyzes chemical libraries by scaffold using Bemis-Murcko scaffolds, generic frameworks, cyclic skeletons, matched molecular pair (MMP) analysis via mmpdb, R-group decomposition, Free-Wilson analysis, scaffold hopping, and chemotype-aware ML train/test splits. Use when identifying chemotype clusters in a library, deriving SAR transformation rules, decomposing series into R-groups, performing scaffold-balanced QSAR splits, or planning analog campaigns.

The English summary is being prepared. Showing the original description.