Improve database description for more reliable skill calling

.claude-plugin/marketplace.json

@@ -7,7 +7,7 @@
   },
   "metadata": {
     "description": "Claude scientific skills from K-Dense Inc",
-    "version": "1.18.0"
+    "version": "1.18.1"
   },
   "plugins": [
     {

scientific-databases/alphafold-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: alphafold-database
-description: Work with the AlphaFold Protein Structure Database to search, retrieve, and analyze AI-predicted protein structures. Use this skill when working with predicted protein structures, UniProt accessions, retrieving confidence scores (pLDDT, PAE), downloading structure files, querying the 200M+ AlphaFold predictions, accessing bulk datasets via Google Cloud, or when needing programmatic access to AlphaFold structural predictions for computational workflows.
+description: Access and analyze AlphaFold protein structure predictions from the DeepMind/EMBL-EBI database containing 200M+ AI-predicted protein structures. Use this skill for: retrieving protein structures by UniProt accession codes, downloading structure files (mmCIF, PDB), accessing confidence metrics (pLDDT scores, PAE matrices), bulk proteome downloads via Google Cloud Storage, structural analysis workflows, drug discovery target preparation, protein engineering studies, evolutionary structural comparisons, structural bioinformatics pipelines, molecular modeling preparation, protein function prediction from structure, structural genomics projects, comparative structural analysis, protein domain identification, binding site analysis, structural quality assessment, confidence score interpretation, batch processing multiple proteins, integrating AlphaFold predictions with experimental data, structural visualization preparation, protein-protein interaction analysis, conformational analysis, structural annotation workflows, homology modeling validation, structural feature extraction, protein classification by structure, structural motif identification, and any computational biology task requiring AI-predicted protein structures.
 ---
 
 # AlphaFold Database

scientific-databases/chembl-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: chembl-database
-description: Toolkit for querying the ChEMBL bioactive molecule database to retrieve compound information, target data, bioactivity measurements, drug mechanisms, and perform structure-based searches. This skill should be used when working with drug discovery data, searching for bioactive compounds, analyzing target-ligand interactions, retrieving pharmaceutical information, or performing cheminformatics queries on small molecules and their biological activities.
+description: Comprehensive toolkit for accessing and querying the ChEMBL database, the world's largest manually curated repository of bioactive drug-like molecules. Use this skill when you need to: search for compounds by name, structure, or molecular properties; retrieve bioactivity data (IC50, Ki, EC50, etc.) for drug targets; find inhibitors, agonists, or bioactive molecules for specific proteins; perform similarity and substructure searches using SMILES; analyze drug-target interactions and mechanisms of action; explore approved drugs and their indications; conduct structure-activity relationship (SAR) studies; identify kinase inhibitors or other drug classes; perform virtual screening based on molecular properties; retrieve pharmaceutical information and drug discovery data; analyze molecular properties like molecular weight, LogP, and drug-likeness; find compounds for drug repurposing studies; query target information for proteins, enzymes, and biological receptors; export bioactivity data for further analysis. This skill is essential for drug discovery, medicinal chemistry, pharmacology, cheminformatics, and any research involving small molecule therapeutics and their biological activities.
 ---
 
 # ChEMBL Database

scientific-databases/clinpgx-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: clinpgx-database
-description: Toolkit for accessing ClinPGx, a clinical pharmacogenomics database providing information on how genetic variation affects drug response. Use this skill when working with pharmacogenomics data, querying gene-drug interactions, accessing CPIC clinical guidelines, retrieving allele function and frequency information, exploring PharmGKB annotations, or conducting research on personalized medicine and precision pharmacotherapy. ClinPGx consolidates PharmGKB, CPIC, and PharmCAT resources.
+description: Comprehensive toolkit for accessing ClinPGx (Clinical Pharmacogenomics Database), the successor to PharmGKB providing clinical pharmacogenomics information on how genetic variation affects drug response, metabolism, efficacy, and toxicity. Use this skill for pharmacogenomics research, clinical decision support, gene-drug interaction queries, CPIC guideline access, allele function and frequency analysis, drug metabolism pathway exploration, personalized medicine implementation, precision pharmacotherapy, adverse drug reaction prediction, genotype-guided dosing, clinical annotation retrieval, drug label analysis, variant interpretation, phenotype prediction, population pharmacogenomics, medication therapy management, clinical trial screening, pharmacogene panel analysis, CYP450 enzyme interactions, transporter gene effects, HLA-associated drug reactions, immunosuppressant dosing, oncology drug toxicity, antidepressant response, anticoagulant therapy, pain medication metabolism, antiviral screening, cardiovascular drug interactions, and PharmDOG clinical decision support tool integration. ClinPGx consolidates PharmGKB, CPIC (Clinical Pharmacogenetics Implementation Consortium), PharmCAT, DPWG guidelines, FDA/EMA drug labels, and provides REST API access to curated pharmacogenomic knowledge for evidence-based clinical practice.
 ---
 
 # ClinPGx Database

scientific-databases/clinvar-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: clinvar-database
-description: Work with ClinVar, NCBI's public archive of genomic variants and their clinical significance. Use when querying variant pathogenicity, searching for disease-associated mutations, interpreting clinical classifications, accessing variant data programmatically via E-utilities API, downloading bulk datasets from FTP, or analyzing variant-condition relationships for research in human genetics and precision medicine.
+description: Access and analyze ClinVar, NCBI's authoritative database of human genomic variants and their clinical significance classifications. Use this skill when: searching for pathogenic, benign, or VUS variants by gene name, chromosome position, or disease condition; interpreting clinical significance classifications (pathogenic, likely pathogenic, uncertain significance, likely benign, benign) and review status star ratings; querying variant pathogenicity for specific genes like BRCA1, BRCA2, TP53, CFTR; accessing ClinVar data programmatically via NCBI E-utilities API (esearch, esummary, efetch); downloading bulk ClinVar datasets from FTP in XML, VCF, or tab-delimited formats; annotating variant call files (VCF) with clinical significance; analyzing variant-condition relationships for genetic disease research; resolving conflicting variant interpretations between submitters; filtering variants by review status (expert panel, practice guidelines); building local ClinVar databases for genomic analysis pipelines; studying hereditary cancer variants, Mendelian disease mutations, or pharmacogenomic variants; performing variant interpretation for precision medicine applications; accessing ClinVar submission data and evidence criteria; tracking variant classification updates over time; or any task requiring authoritative clinical variant interpretation data from NCBI's ClinVar database.
 ---
 
 # ClinVar Database

scientific-databases/cosmic-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: cosmic-database
-description: Work with COSMIC (Catalogue of Somatic Mutations in Cancer), the world's largest database of somatic cancer mutations. Use this skill when working with cancer genomics data, downloading mutation databases, querying cancer gene information, analyzing mutational signatures, or accessing curated cancer gene census data. Applies to research involving cancer mutations, drug resistance, structural variants, copy number alterations, and gene expression in cancer.
+description: Access and analyze COSMIC (Catalogue of Somatic Mutations in Cancer), the world's largest database of somatic cancer mutations. Use this skill for downloading cancer mutation datasets, accessing the Cancer Gene Census, retrieving mutational signatures, analyzing drug resistance mutations, working with structural variants and gene fusions, accessing copy number alterations, and integrating cancer genomics data into bioinformatics pipelines. Essential for cancer research, oncology drug discovery, precision medicine, tumor profiling, biomarker identification, cancer genomics analysis, somatic variant annotation, mutational signature analysis, cancer gene prioritization, drug resistance studies, and cancer cell line research. Supports both academic (free) and commercial (licensed) access with authentication required.
 ---
 
 # COSMIC Database

scientific-databases/ena-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ena-database
-description: Toolkit for accessing and retrieving data from the European Nucleotide Archive (ENA), including programmatic API access for sequences, assemblies, samples, studies, and reads. Use this skill when working with nucleotide sequence data, submitting or retrieving genomic/transcriptomic data, searching for sequence records, or building bioinformatics pipelines that require ENA data access.
+description: Comprehensive toolkit for accessing, searching, and retrieving data from the European Nucleotide Archive (ENA) - the primary European repository for nucleotide sequence data. Provides programmatic API access for DNA/RNA sequences, genome assemblies, raw sequencing reads (FASTQ), samples, studies, experiments, runs, analyses, and taxonomic records. Use this skill for: retrieving genomic/transcriptomic data by accession numbers (ERR, SRR, PRJ, GCA, etc.), searching sequence databases, downloading raw sequencing data, accessing genome assemblies, finding samples and studies, building bioinformatics pipelines, performing sequence similarity searches, accessing taxonomic information, bulk data downloads, metadata extraction, and integrating ENA data into computational biology workflows. Supports multiple data formats (FASTQ, FASTA, BAM, CRAM, XML, JSON, TSV) and download methods (API, FTP, Aspera). Essential for genomics, transcriptomics, metagenomics, phylogenetics, and molecular biology research requiring access to European nucleotide sequence repositories.
 ---
 
 # ENA Database

scientific-databases/ensembl-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ensembl-database
-description: Work with the Ensembl genome database to query genomic data, retrieve sequences, analyze variants, and perform comparative genomics. This skill should be used when working with vertebrate genomic data, gene annotations, variant analysis, ortholog identification, or when users need to query the Ensembl REST API for genomic information across multiple species.
+description: Access and query the Ensembl genome database for comprehensive vertebrate genomic data analysis. Use this skill for gene lookups, sequence retrieval, variant analysis, comparative genomics, ortholog/paralog identification, genomic region analysis, assembly mapping, and VEP predictions. Handles gene symbols, Ensembl IDs, rsIDs, genomic coordinates, chromosome regions, cross-species comparisons, evolutionary analysis, regulatory elements, transcript/protein sequences, population genetics, phenotype associations, and genome assembly conversions. Supports REST API queries across 250+ species including human, mouse, zebrafish, and other vertebrates. Essential for genomics research, variant interpretation, evolutionary studies, gene annotation pipelines, and bioinformatics workflows requiring authoritative genomic reference data.
 ---
 
 # Ensembl Database

scientific-databases/gene-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: gene-database
-description: Work with NCBI Gene database to search, retrieve, and analyze gene information including nomenclature, sequences, variations, phenotypes, and pathways. This skill should be used when querying gene databases, looking up gene symbols or IDs, retrieving gene sequences or metadata, analyzing gene functions, or accessing NCBI Gene programmatically using E-utilities or Datasets API.
+description: Access and query NCBI Gene database programmatically using E-utilities and Datasets API. Search genes by symbol, name, ID, or biological context across organisms. Retrieve comprehensive gene information including nomenclature, aliases, reference sequences (RefSeqs), chromosomal locations, Gene Ontology annotations, phenotypes, pathways, and cross-references. Perform batch gene lookups, validate gene lists, analyze gene functions, and access gene metadata. Handle gene symbol resolution, organism-specific queries, and gene identifier mapping. Use for gene annotation, functional analysis, pathway enrichment, variant interpretation, and genomic data integration workflows. Supports JSON, XML, GenBank, FASTA, and text output formats with rate limiting and error handling.
 ---
 
 # Gene Database

scientific-databases/geo-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: geo-database
-description: Work with the Gene Expression Omnibus (GEO) database to search, retrieve, and analyze high-throughput gene expression and functional genomics data. Use this skill when working with microarray data, RNA-seq datasets, gene expression profiles, GEO accessions (GSE, GSM, GPL, GDS), downloading SOFT/MINiML files, querying expression experiments, performing differential expression analysis, accessing GEO metadata, or when needing programmatic access to functional genomics data repositories.
+description: Work with the Gene Expression Omnibus (GEO) database to search, retrieve, download, and analyze high-throughput gene expression and functional genomics data. Use this skill for microarray data analysis, RNA-seq datasets, gene expression profiling, accessing GEO accessions (GSE series, GSM samples, GPL platforms, GDS datasets), downloading SOFT/MINiML/Matrix files, querying expression experiments, performing differential expression analysis, accessing GEO metadata, batch processing multiple datasets, quality control of expression data, correlation analysis, clustering, meta-analysis across studies, biomarker discovery, drug response studies, disease biology research, transcriptomics analysis, or when needing programmatic access to functional genomics repositories. This skill covers GEOparse library usage, NCBI E-utilities API, FTP downloads, data preprocessing, statistical analysis, visualization, and integration with downstream analysis workflows.
 ---
 
 # GEO Database

scientific-databases/gwas-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: gwas-database
-description: Toolkit for querying the GWAS Catalog, the NHGRI-EBI database of published genome-wide association studies. Use this skill when searching for genetic variants associated with diseases or traits, retrieving SNP-trait associations, accessing GWAS summary statistics, exploring gene-disease relationships, conducting genetic epidemiology research, or performing systematic reviews of genetic associations. Essential for human genetics, precision medicine, and genomic research.
+description: Comprehensive toolkit for accessing and querying the GWAS Catalog (NHGRI-EBI database of published genome-wide association studies). Use this skill when you need to: find genetic variants (SNPs) associated with diseases or traits, retrieve SNP-trait associations with p-values and effect sizes, access GWAS summary statistics and genome-wide data, explore gene-disease relationships and pleiotropy, conduct genetic epidemiology research, perform systematic reviews of genetic associations, identify variants for polygenic risk scores, investigate population-specific genetic associations, access curated SNP-trait associations from thousands of GWAS publications, query by rs ID variants, search by disease/trait names, find variants in specific genes or chromosomal regions, retrieve study metadata and publication information, access full summary statistics for downstream analysis, cross-reference with Ensembl and other genomic databases, or perform meta-analyses of genetic associations. Essential for human genetics research, precision medicine applications, genomic research, pharmacogenomics, population genetics, and genetic risk prediction studies.
 ---
 
 # GWAS Catalog Database

scientific-databases/hmdb-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: hmdb-database
-description: Work with the Human Metabolome Database (HMDB) for metabolomics research, biomarker discovery, and metabolite identification. Use this skill when searching for metabolite information, querying human metabolic pathways, identifying biomarkers, analyzing metabolomics data, or accessing metabolite properties, structures, and clinical associations. The skill covers web-based searches, downloadable datasets, and programmatic access methods.
+description: Access and analyze the Human Metabolome Database (HMDB) for comprehensive metabolite information, metabolomics research, biomarker discovery, metabolite identification, pathway analysis, and clinical associations. Use this skill for: searching metabolites by name, HMDB ID, structure, or spectral data; retrieving chemical properties (SMILES, InChI, molecular weight, formula); accessing clinical biomarker data and disease associations; downloading bulk datasets (XML, SDF, CSV formats); analyzing metabolic pathways and enzyme associations; performing spectral matching for metabolite identification; accessing concentration ranges in biological fluids; integrating with external databases (KEGG, PubChem, ChEBI); and supporting untargeted metabolomics workflows. HMDB contains 220,945+ metabolite entries with chemical, biological, clinical, and analytical data including NMR/MS spectra, pathway information, and biomarker associations for human metabolomics research.
 ---
 
 # HMDB Database

scientific-databases/kegg-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: kegg-database
-description: Work with the KEGG (Kyoto Encyclopedia of Genes and Genomes) database for biological pathway analysis, gene-to-pathway mapping, compound searches, and molecular interaction networks. Use this skill when working with pathway enrichment analysis, metabolic pathways, gene annotations, drug-drug interactions, or converting between biological database identifiers (KEGG, UniProt, NCBI Gene, PubChem). Supports querying pathways, genes, compounds, enzymes, diseases, and drugs across multiple organisms.
+description: Access and analyze the KEGG (Kyoto Encyclopedia of Genes and Genomes) database for comprehensive biological pathway analysis, molecular interaction networks, and cross-database integration. Use this skill for pathway enrichment analysis, gene-to-pathway mapping, metabolic pathway exploration, drug-drug interaction checking, compound structure retrieval, enzyme pathway analysis, disease pathway investigation, and identifier conversion between KEGG and external databases (UniProt, NCBI Gene, PubChem, ChEBI). Supports querying pathways, genes, compounds, enzymes, diseases, drugs, reactions, modules, and orthology groups across multiple organisms including human, mouse, yeast, E. coli, and fruit fly. Key operations include database information retrieval, entry listing and searching, detailed entry retrieval in multiple formats (FASTA sequences, MOL structures, pathway images, KGML XML, JSON), cross-referencing between databases, ID conversion, and drug interaction analysis. Essential for bioinformatics workflows involving pathway analysis, systems biology, drug discovery, metabolic engineering, comparative genomics, and functional annotation of genes and proteins.
 ---
 
 # KEGG Database

scientific-databases/metabolomics-workbench-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: metabolomics-workbench-database
-description: Toolkit for accessing and querying the Metabolomics Workbench, an NIH-sponsored repository containing 4,200+ metabolomics studies with standardized nomenclature (RefMet), study metadata, experimental results, and comprehensive metabolite databases. Use this skill when working with metabolomics data, querying metabolite structures, accessing study results, standardizing metabolite names, performing mass spectrometry searches, or retrieving gene/protein associations with metabolites.
+description: Comprehensive toolkit for accessing and analyzing metabolomics data through the Metabolomics Workbench REST API. This NIH-sponsored repository contains 4,200+ metabolomics studies with standardized RefMet nomenclature, experimental datasets, metabolite structures, and gene/protein associations. Use this skill for: querying metabolite structures and downloading molecular data (MOL files, PNG images), accessing study metadata and experimental results from GC-MS/LC-MS/NMR platforms, standardizing metabolite names using RefMet classification system, performing mass spectrometry searches by m/z values with ion adducts, filtering studies by analytical methods (LCMS/GCMS/NMR), ionization polarity (positive/negative), chromatography types (HILIC/RP/GC), species, sample sources, and diseases, retrieving gene and protein information related to metabolic pathways, cross-referencing metabolite identifiers across databases (PubChem, KEGG, HMDB), identifying compounds from MS data using exact mass calculations, exploring disease-specific metabolomics studies, accessing untargeted metabolomics datasets, and retrieving complete experimental data in JSON or TXT formats. Essential for metabolomics research, biomarker discovery, metabolic pathway analysis, and mass spectrometry data interpretation.
 ---
 
 # Metabolomics Workbench Database

scientific-databases/pdb-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: pdb-database
-description: Work with the RCSB Protein Data Bank (PDB) to search, retrieve, and analyze 3D structures of proteins, nucleic acids, and other biological macromolecules. Use this skill when working with protein structures, PDB IDs, crystallographic data, protein structure analysis, molecular visualization, structure-function relationships, or when needing to query or download structural biology data programmatically.
+description: Access and analyze the RCSB Protein Data Bank (PDB) - the global repository for 3D structural data of biological macromolecules including proteins, nucleic acids, complexes, and ligands. This skill enables searching structures by text, attributes, sequence similarity, and structural similarity; retrieving detailed metadata and experimental information; downloading coordinate files in PDB, mmCIF, and BinaryCIF formats; performing batch operations on multiple structures; and integrating PDB data into computational workflows. Use this skill for protein structure analysis, molecular visualization, drug discovery research, protein engineering, structural biology studies, crystallographic data analysis, homology modeling, ligand binding site analysis, structure-function relationship studies, evolutionary analysis, educational content creation, and any task requiring access to experimentally determined or computationally predicted macromolecular structures. Key capabilities include querying by organism, resolution, experimental method, deposition date, biological assembly information, and performing sequence/structure similarity searches across the entire PDB archive.
 ---
 
 # PDB Database

scientific-databases/pubchem-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: pubchem-database
-description: Access chemical compound data from PubChem, the world's largest free chemical database. This skill should be used when retrieving compound properties, searching for chemicals by name/SMILES/InChI, performing similarity or substructure searches, accessing bioactivity data, converting between chemical formats, or generating chemical structure images. Works with over 110 million compounds and 270 million bioactivities through PUG-REST API and PubChemPy library.
+description: Access and analyze chemical compound data from PubChem database using PubChemPy and PUG-REST API. Use this skill when you need to: search compounds by name/CID/SMILES/InChI/formula, retrieve molecular properties (MW/LogP/TPSA/H-bond counts), perform similarity searches with Tanimoto thresholds, conduct substructure searches for pharmacophores, convert between chemical formats (SMILES/InChI/SDF/JSON), generate 2D structure images, access bioactivity data from assays, get compound synonyms and annotations, screen compounds using Lipinski's Rule of Five, batch process multiple compounds, or find drug-like candidates. Handles 110M+ compounds and 270M+ bioactivities with rate limiting (5 req/sec, 400 req/min). Includes error handling for timeouts, not found errors, and missing properties. Supports both synchronous and asynchronous operations for large similarity/substructure searches.
 ---
 
 # PubChem Database

scientific-databases/pubmed-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: pubmed-database
-description: Search and retrieve biomedical research articles from PubMed. Use when users need to find medical/scientific papers, search for literature on diseases/treatments/drugs, get research articles by author/topic/PMID, conduct literature reviews, construct advanced search queries, or access biomedical publications programmatically. Covers medicine, biology, pharmacology, genetics, and all life sciences research.
+description: Comprehensive PubMed database expertise for searching, retrieving, and analyzing biomedical research literature. Use for literature searches, systematic reviews, meta-analyses, citation discovery, programmatic data access, and biomedical research workflows. Handles advanced search queries with Boolean operators, MeSH terms, field tags, publication type filters, and date ranges. Provides E-utilities API integration for automated workflows, batch processing, and large-scale data extraction. Supports citation management, export formats, search history, and related article discovery. Covers medicine, biology, pharmacology, genetics, clinical trials, epidemiology, drug discovery, disease research, treatment protocols, and all life sciences domains. Essential for researchers, clinicians, students, and anyone conducting biomedical literature analysis or evidence-based research.
 ---
 
 # PubMed Database

scientific-databases/reactome-database/reactome-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: reactome-database
-description: Work with Reactome pathway database for analyzing biological pathways, performing pathway enrichment analysis, querying molecular interactions, and analyzing gene expression data. This skill should be used when working with biological pathways, performing overrepresentation analysis, mapping gene identifiers to pathways, analyzing gene expression datasets, or exploring disease-related pathways. Supports both direct REST API access and the reactome2py Python package.
+description: Comprehensive Reactome pathway database integration for biological pathway analysis, enrichment studies, molecular interaction queries, and gene expression analysis. Use this skill for pathway overrepresentation analysis, gene-to-pathway mapping, expression dataset analysis, disease pathway exploration, molecular interaction networks, pathway hierarchy queries, species comparison studies, pathway visualization, and statistical pathway enrichment. Supports REST API access to Content Service (data retrieval) and Analysis Service (computational analysis), plus reactome2py Python package integration. Handles gene symbols, UniProt IDs, Ensembl IDs, EntrezGene IDs, ChEBI IDs, and expression data formats. Provides pathway statistics, literature references, molecular entities, reactions, complexes, and pathway browser visualization links.
 ---
 
 # Reactome Database

scientific-databases/string-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: string-database
-description: Work with the STRING (Search Tool for the Retrieval of Interacting Genes/Proteins) database for protein-protein interaction network analysis, functional enrichment, and interaction partner discovery. Use this skill when analyzing protein interactions, building protein networks, performing pathway enrichment on protein sets, finding interaction partners, testing network connectivity significance, or retrieving protein homology data. Covers 5000+ genomes with 59.3 million proteins and 20+ billion interactions from experimental data, computational prediction, and text-mining.
+description: Access and analyze the STRING database for comprehensive protein-protein interaction (PPI) network analysis, functional enrichment, pathway analysis, and protein interaction discovery. This skill enables querying protein interactions, building interaction networks, performing Gene Ontology (GO) enrichment, KEGG pathway analysis, Pfam domain enrichment, protein-protein interaction enrichment testing, network visualization, interaction partner discovery, homology analysis, and identifier mapping across 5000+ species. Use for analyzing protein lists from experiments (differential expression, proteomics, mass spectrometry), validating protein networks, discovering novel protein interactions, pathway enrichment analysis, functional annotation, network connectivity analysis, cross-species protein comparison, protein family analysis, hub protein identification, network expansion from seed proteins, and systems biology studies. Provides access to 59.3 million proteins and 20+ billion interactions from experimental data, computational predictions, text-mining, and curated databases. Supports confidence scoring, multiple evidence types (experimental, coexpression, phylogenetic, genomic context), physical vs functional networks, and visualization capabilities.
 ---
 
 # STRING Database

scientific-databases/uniprot-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: uniprot-database
-description: Work with the UniProt protein database for retrieving protein sequences, annotations, and functional information. Use this skill when searching for protein data, retrieving protein sequences (FASTA format), accessing protein annotations, mapping protein identifiers, querying UniProtKB/Swiss-Prot or TrEMBL databases, or performing batch protein queries.
+description: Access and query the UniProt protein database for comprehensive protein information retrieval. Use this skill when you need to search for proteins by name, gene symbol, accession number, or functional terms; retrieve protein sequences in FASTA format; access detailed protein annotations including function, structure, interactions, and pathways; map protein identifiers between different databases (Ensembl, RefSeq, PDB, KEGG, GO terms); query Swiss-Prot (reviewed) or TrEMBL (unreviewed) protein entries; perform batch operations on multiple proteins; download protein datasets; analyze protein families and domains; investigate protein-protein interactions; explore evolutionary relationships through UniRef clusters; access protein structure predictions from AlphaFoldDB; retrieve Gene Ontology annotations; analyze protein modifications and post-translational sites; or work with protein sequences for bioinformatics analysis. This skill provides programmatic access to UniProtKB, UniRef, UniParc, and related databases through REST API endpoints with support for various output formats (JSON, TSV, FASTA, XML).
 ---
 
 # UniProt Database

scientific-databases/zinc-database/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: zinc-database
-description: Toolkit for accessing the ZINC database of commercially-available compounds for virtual screening and drug discovery. This skill should be used when searching for purchasable molecules, performing structure-based searches, retrieving compounds for molecular docking, exploring chemical space, or querying compounds by ZINC ID, SMILES, supplier codes, or molecular properties for lead discovery and virtual screening campaigns.
+description: Access and query the ZINC database containing 230+ million commercially-available compounds for virtual screening, drug discovery, and molecular docking studies. Use this skill when you need to: search for purchasable molecules by ZINC ID, SMILES, or supplier codes; perform structural similarity searches and analog discovery; retrieve random compound sets for screening libraries; find lead compounds for drug development; explore chemical space for virtual screening campaigns; download 3D-ready molecular structures for docking; verify compound availability from chemical suppliers; perform batch compound retrieval; generate screening libraries based on drug-likeness criteria; find commercially-available analogs of known active compounds; access the CartBlanche22 API for programmatic compound searches; filter compounds by molecular properties (LogP, MW, H-bond donors); retrieve compounds from specific catalogs or vendors; perform substructure searches; generate diverse compound sets for high-throughput screening; find fragment-like molecules for fragment-based drug discovery; access ready-to-dock 3D conformations; cross-reference supplier information; perform chemical space sampling; and integrate ZINC data with molecular docking workflows. This skill provides both web interface access and API endpoints for automated compound discovery and virtual screening applications.
 ---
 
 # ZINC Database