Context & AI engineering
25,772 rows, by stacks then stars
1,865 of these skill files have been read, by 1,328 distinct authors, and what they tell an agent counted. What Context & AI engineering authors agree on, and what they forbid
Embedding space representation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/embedding-space-representation Skill
16★ repoUse when you have pairs or triplets of MS/MS spectra with associated metadata (compound structural information, Tanimoto similarity scores) and want to learn embeddings that simultaneously preserve spectral similarity relationships and reconstruct peak intensities.
Embedding space visualization and interpretation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/embedding-space-visualization-and-interpretation Skill
16★ repoUse when after training or loading a pre-trained deep learning model that produces high-dimensional embeddings (e.g., 200-dimensional vectors from MS2DeepScore). Use this skill when you have a set of spectra with known chemical class annotations (e.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/embedding-vector-generation Skill
16★ repoUse when when you have tokenized mass spectra (peak-mass and peak-intensity pairs from experimental or in-silico libraries such as NIST 2017 or MassBank) and need to perform rapid similarity searches or spectrum matching at scale.
Embedding vector similarity ranking
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/embedding-vector-similarity-ranking Skill
16★ repoUse when you have a query mass spectrum and need to retrieve the most chemically similar reference spectra from a large experimental or predicted library (thousands to millions of entries).
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/embedding-vector-validation Skill
16★ repoUse when after instantiating and invoking a sinusoidal formula embedding layer (such as SCARF embeddings in MIST-CF) on chemical formula inputs, validate that the output embeddings meet dimensionality and value constraints before using them for downstream transformer or ranking tasks.
Few shot learning calibration design
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/few-shot-learning-calibration-design Skill
16★ repoUse when you have experimental retention times measured on a source chromatographic method and want to predict RTs on a target chromatographic method, but possess only a small set (10–100) of molecules with ground-truth measurements on both methods.
Gallery benchmark data extraction
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/gallery-benchmark-data-extraction Skill
16★ repoUse when you have access to a computation-times table or performance log documenting rendering execution times for multiple visualization examples across different plotting backends (e.
Graph embedding coordinate extraction
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/graph-embedding-coordinate-extraction Skill
16★ repoUse when you have a network object (loaded as igraph or MetaNet format with edges and nodes defined) and need to assign two-dimensional spatial coordinates to nodes for downstream visualization on static platforms (ggplot2, base R graphics) or interactive viewers (Gephi, Cytoscape).
Heterogeneous graph embedding design
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/heterogeneous-graph-embedding-design Skill
16★ repoUse when when building a Graph Transformer model for continuous property prediction on molecules with associated experimental or instrumental metadata (e.g., retention time prediction across different chromatographic columns, methods, or conditions).
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/hpc-job-array-orchestration Skill
16★ repoUse when you have a machine learning training workflow (e.g., k-fold cross-validation) where each fold is independent, GPU-accelerated, and can run in parallel.
In memory cache implementation for repeated queries
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/in-memory-cache-implementation-for-repeated-queries Skill
16★ repoUse when when deploying a high-throughput molecular classification API (e.g., /classify endpoint) where the same SMILES strings are expected to be queried repeatedly across multiple users or time windows, and response latency is a performance constraint.
In memory optimization for large datasets
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/in-memory-optimization-for-large-datasets Skill
16★ repoUse when you are repeatedly querying or iterating over multidimensional MS data stored in MZA HDF5 format (retention time, drift time, m/z dimensions) and profiling shows that repeated disk I/O for the same metadata or scan ranges dominates runtime.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/interactive-plot-embedding Skill
16★ repoUse when you have resolved USI (Unified Spectrum Identifier) spectrum data from a supported repository (GNPS, MassBank, MetaboLights, Metabolomics Workbench, ProteoXchange, or MS2LDA) and need to create a figure suitable for journal publication or supplementary materials that retains a link to the.
Ion image embedding optimization
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/ion-image-embedding-optimization Skill
16★ repoUse when you have 512-dimensional representation vectors output from ResNet18 encoders processing paired augmented ion images, and you need to prevent trivial solutions (representation collapse) during contrastive learning—specifically when optimizing for maximized similarity between augmentations.
Java maven build orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/java-maven-build-orchestration Skill
16★ repoUse when when you have access to a Java project with a Maven pom.xml defining multiple interdependent modules (e.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/joint-embedding-space-scoring Skill
16★ repoUse when you have a mass spectrum from an untargeted metabolomics experiment and a set of candidate molecules (e.g., downloaded from PubChem) that may explain that spectrum. You want to rank these candidates by likelihood of correctness to prioritize manual annotation or further validation.
Jvm heap memory tuning verification
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/jvm-heap-memory-tuning-verification Skill
16★ repoUse when when deploying the ipbhalle/metfragweb Docker container and you need to confirm that custom JVM heap sizes (specified via JAVA_OPTS=-Xmx?g -Xms?g) have taken effect, or when troubleshooting memory-related runtime issues in the running container.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/knime-workflow-orchestration Skill
16★ repoUse when you have raw LC-MS data (mzML, NetCDF) from multiple runs that require sequential feature detection, alignment, quantification, and optional filtering (e.g., blank exclusion, QC reproducibility, retention-time outlier removal) before spatial mapping or annotation.
Laplacian embedding spectral alignment
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/laplacian-embedding-spectral-alignment Skill
16★ repoUse when you have two or more MS/MS fragmentation spectra (with precursor m/z, fragment m/z values, and intensities) and need to identify which fragment ions match across spectra while quantifying the statistical confidence of those matches.
Lc ms scan acquisition orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/lc-ms-scan-acquisition-orchestration Skill
16★ repoUse when when you have a curated list of chemical compounds (real or virtual), a defined fragmentation strategy (e.g., Top-N, exclusion lists), and need to simulate how that strategy will acquire MS1 and MS2 scans over a defined retention-time window.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/lcms-workflow-orchestration Skill
16★ repoUse when starting from raw LC-MS spectral files (mzML or mzXML format) in a global metabolomics study and you need to produce a complete, validated feature table with m/z, retention time, and intensity values across all samples.
Mass spectra embedding extraction
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectra-embedding-extraction Skill
16★ repoUse when you have tandem mass spectra (MS/MS) in .msp format and need dense, chemically meaningful vector representations for library matching, similarity computation, or structural clustering. Apply this when comparing spectra across large reference databases (e.
Mass spectral library benchmark execution
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectral-library-benchmark-execution Skill
16★ repoUse when you have access to the Flash Entropy Search implementation and want to validate that real-time spectral library querying achieves the reported timing and throughput metrics under the same library sizes, query parameters, and measurement methodology documented in the paper.
Mass spectrometry benchmark analysis
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectrometry-benchmark-analysis Skill
16★ repoUse when you have implemented or modified a tandem mass spectrometry formula inference model and need to measure whether a specific architectural change (e.
Mass spectrometry benchmark design
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectrometry-benchmark-design Skill
16★ repoUse when when claiming that one mass spectrometry processing library achieves higher throughput than competitors, or when evaluating whether a new or optimized implementation delivers the expected computational efficiency gains.
Mass spectrometry embedding generation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectrometry-embedding-generation Skill
16★ repoUse when when you have preprocessed MS/MS spectral data (normalized peak intensities and m/z values) and need to convert individual spectra into fixed-dimensional vector representations for similarity-based metabolite matching or comparative analysis.
Mass spectrometry spectral embedding learning
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectrometry-spectral-embedding-learning Skill
16★ repoUse when when you have preprocessed MS/MS spectral pairs (peak intensities and m/z values) and need to predict molecular structural similarity scores, or when you want to project spectra into a learned chemical embedding space for visualization (e.g., via UMAP) or downstream similarity searches.
Mass spectrometry workflow orchestration snakemake
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectrometry-workflow-orchestration-snakemake Skill
16★ repoUse when when you have a collection of mzML.gz files from a multidimensional MS instrument (e.
Mass spectrometry workflow orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectrometry-workflow-orchestration Skill
16★ repoUse when you have a collection of mzML or mzML.gz files from LC-IMS-MS/MS experiments and need to apply a consistent, reproducible sequence of feature detection, alignment, CCS calibration, isotope detection, and MS/MS deconvolution operations across multiple samples.
Mass spectrum embedding generation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/mass-spectrum-embedding-generation Skill
16★ repoUse when you have cleaned MS/MS spectra (in formats like .mgf, .msp, .
Memory efficient pairwise statistics
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/memory-efficient-pairwise-statistics Skill
16★ repoUse when you have an MSI intensity matrix (msimat object) and an annotated mass-difference table (massdiff object with known or hypothesized parent–adduct ion pairs) where the number of peak pairs exceeds available RAM.
Memory mapped file selection strategy
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/memory-mapped-file-selection-strategy Skill
16★ repoUse when initializing a Dataset object in NMRFx and must decide which storage backend to use for in-memory or memory-mapped file access.
Memory profiling and benchmarking
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/memory-profiling-and-benchmarking Skill
16★ repoUse when when designing or optimizing backends that handle large MS datasets (mzML, mzXML, CDF files via MsBackendMzR), to verify that claimed memory advantages of on-disk or chunked approaches actually materialize in practice.
Metabolite benchmark dataset validation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/metabolite-benchmark-dataset-validation Skill
16★ repoUse when after mzRAPP has exported a benchmark CSV file from processing a batch of centroided mzML files (e.
Metabolite benchmark peak matching
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/metabolite-benchmark-peak-matching Skill
16★ repoUse when you have LC-HRMS mzML files processed by a non-targeted peak-detection tool (XCMS, MZmine 2/3, MS-DIAL, OpenMS, etc.) and want to quantify how many of a known set of target molecules (47+ compounds with isotopologues) were correctly detected, aligned, and preserved with intact isotopologue.
Metabolite database embedding lookup
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/metabolite-database-embedding-lookup Skill
16★ repoUse when when you have paired tandem MS spectra and known molecular structures (SMILES or fingerprints) and want to annotate novel spectra by retrieving similar structures from a reference database without relying on spectral database matching.
Molecular embedding generation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/molecular-embedding-generation Skill
16★ repoUse when when you have molecular structures (SMILES or explicit graph representations) and need fixed-size vector representations to feed into a prediction head or transfer learning task.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/molecular-embedding-inference Skill
16★ repoUse when you have a set of molecules (as SMILES strings in CSV format) and a pre-trained NaFM checkpoint, and you need to generate embeddings or predictions for virtual screening, bioactivity regression, or compound ranking tasks.
Molecular model benchmark comparison
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/molecular-model-benchmark-comparison Skill
16★ repoUse when when you have (1) a candidate foundation model or pre-trained weights for natural products (e.g., NaFM), (2) one or more baseline models pre-trained on synthetic molecules (e.g., ChemBERTa, MolBERT), (3) a held-out test set from a downstream task (e.
Multidimensional scaling embedding
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/multidimensional-scaling-embedding Skill
16★ repoUse when after computing a pairwise sample distance matrix from aligned MS2 fingerprint vectors and you need to visualize sample relationships, clustering, or separation by group identity in 2D space.
Natural products workflow orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/natural-products-workflow-orchestration Skill
16★ repoUse when you have LC-MS/MS DDA metabolomics data (positive and/or negative ionization modes) and sample metadata (originating taxon) for one or more samples, and you need to generate a Wikidata-connected RDF knowledge graph for integrated natural products analysis, taxonomy-aware compound.
Neural network inference and embedding extraction
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/neural-network-inference-and-embedding-extraction Skill
16★ repoUse when when you have paired or unpaired MS/MS spectra and need to compute structural similarity scores without explicit molecular fingerprint computation, or when you want to generate low-dimensional embeddings for spectral visualization, clustering, or retrieval tasks.
Nextflow pipeline orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/nextflow-pipeline-orchestration Skill
16★ repoUse when when you have raw LC-HRMS metabolomics data in .mzML or .
Nextflow workflow orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/nextflow-workflow-orchestration Skill
16★ repoUse when when you need to coordinate multiple bioinformatics tools (quality control, alignment, quantification, normalization, batch correction) across diverse compute environments (HPC, cloud, local) while ensuring reproducibility and parameter traceability.
Orchestrator architecture design
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/orchestrator-architecture-design Skill
16★ repoUse when when building a multi-backend visualization library where users specify both a plot type (spectrum, chromatogram, peakmap) and a backend (matplotlib for static output, Bokeh or Plotly for interactive), and you need to avoid code duplication across backends while keeping the user-facing API.
Parallel computing workflow orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/parallel-computing-workflow-orchestration Skill
16★ repoUse when when compiling EI or MS/MS spectral libraries from multiple gigabyte-scale sources (e.
Performance benchmark execution
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/performance-benchmark-execution Skill
16★ repoUse when you have access to a tool with published performance claims (e.g., '~20 seconds on M2 Macbook') and want to validate those claims on your own hardware, or you need to benchmark processing speed before committing to a tool for production use on similarly sized datasets.
Python workflow execution and orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/python-workflow-execution-and-orchestration Skill
16★ repoUse when you have raw LC/MS data in mzML format and need to execute a complete non-targeted screening workflow to extract and annotate chemical features.
Qr code generation and embedding
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/qr-code-generation-and-embedding Skill
16★ repoUse when you need to publish metabolomics spectra in static media (PDF, print, supplementary tables) and want readers or automated systems to access the corresponding interactive spectrum visualization without manual lookup.
R bioconductor workflow orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/r-bioconductor-workflow-orchestration Skill
16★ repoUse when you have untargeted LC/MS metabolomics data from stable isotope labeling experiments (e.g., 13C-glucose vs. 12C-glucose) already converted to mzXML format, and you need to systematically identify putatively incorporated metabolic features by comparing unlabeled and labeled sample groups.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/r-function-calling Skill
16★ repoUse when after AutoTuner has completed peak identification (TIC analysis), peak isolation, and EIC parameter extraction on at least 3 raw mass spectrometry samples (qTOF, orbitrap, or FTICR formats converted to mzML/mzXML/CDF), and you need to export the tuned parameters in a format ready for XCMS.
Rna seq preprocessing pipeline orchestration
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/rna-seq-preprocessing-pipeline-orchestration Skill
16★ repoUse when when you have raw FASTQ files (from SRA or local storage) and need to produce normalized transcript quantification (quant.sf) files for a multi-sample RNA-seq cohort.
S4 class object memory profiling
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/s4-class-object-memory-profiling Skill
16★ repoUse when when designing or optimizing S4-based data backends (such as MsBackend subclasses) and you need to decide whether to pre-populate all slots with complete data structures or defer initialization until data access.
Siamese network embedding generation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/siamese-network-embedding-generation Skill
16★ repoUse when you have preprocessed MS/MS spectra binned into 10,000 equally-sized m/z bins (10–1000 m/z range) with square-root-transformed intensities, and you need to generate 200-dimensional spectral embeddings for structural similarity prediction, visualization via dimensionality reduction (e.
Sinusoidal embedding implementation
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/sinusoidal-embedding-implementation Skill
16★ repoUse when building a transformer-based neural network for chemical formula ranking or classification from mass spectrometry spectra, and you need to encode categorical chemical formulas (e.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/spatial-feature-embedding Skill
16★ repoUse when when analyzing imaging mass spectrometry datasets where you need to reduce high-dimensional peak intensity features while preserving spatial structure, and when automatic peak picking and marker ion identification are required.
Spectral embedding extraction from neural networks
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/spectral-embedding-extraction-from-neural-networks Skill
16★ repoUse when you have a set of preprocessed MS/MS spectra (binned to 10,000 m/z bins with intensities square-root transformed) and a trained MS2DeepScore Siamese neural network model, and you need to compute 200-dimensional spectral embeddings for dimensionality reduction, clustering, or chemical space.
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/spectral-embedding-generation Skill
16★ repoUse when you have a collection of pre-processed MS/MS spectra (binned, intensity-normalized) and a trained MS2DeepScore base network, and you need to compute structural similarity scores between spectrum pairs or visualize spectra in chemical space via dimensionality reduction (e.g., UMAP).
Spectral embedding rationality verification
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/spectral-embedding-rationality-verification Skill
16★ repoUse when after training a transformer-encoder-based mass spectrometry embedding model (e.
Spectral molecular embedding alignment
HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/spectral-molecular-embedding-alignment Skill
16★ repoUse when you have MS/MS spectra in MGF or similar format and a reference library of molecular structures (SMILES or SDF), and your goal is to retrieve the most likely structures for an unknown compound spectrum by learning a joint embedding space rather than using traditional spectral similarity or.