K-Dense-AI/scientific-agent-skills

Adaptyv Bio Foundry API connects agents to the Adaptyv Bio cloud lab for submitting protein sequences, running assays, and retrieving experimental results. Use it when building workflows for protein binding, thermostability, expression, or fluorescence experiments.

Aeon is a scikit-learn compatible toolkit for time series machine learning, covering classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use it when working with temporal data or time-indexed observations requiring specialized algorithms beyond standard ML approaches.

Analytical Method Validation plans and evaluates validation, verification, and transfer of analytical procedures under ICH Q2(R2), Q14, USP <1225>, CLSI EP, and ISO/IEC 17025. Use it for HPLC, LC-MS/MS, GC, qPCR, and ligand binding assays.

AnnData handles annotated data matrices for single-cell analysis, storing measurements, metadata, and multi-dimensional annotations. Use it when working with .h5ad files or integrating with the scverse ecosystem.

Arbor autonomously improves a real artifact (code, training recipe, agent harness, data pipeline, prompt) against an objective and an evaluator using Hypothesis Tree Refinement (HTR). Use it for iterative optimization over many experiments without overfitting.

Arboreto infers gene regulatory networks from gene expression data using scalable ensemble regression algorithms (GRNBoost2, GENIE3). Use when analyzing bulk or single-cell RNA-seq data to identify transcription factor-target relationships and regulatory interactions.

Astropy provides core Python functionality for astronomy and astrophysics workflows, including units, coordinates, FITS I/O, tables, time, WCS, and cosmology. Use when implementing or debugging astronomical data analysis code with Astropy.

Autoskill observes the user's local screenpipe timeline to detect repeated research workflows, matches them against existing skills, and drafts new skill proposals or composition recipes for uncovered patterns. Use it when the user explicitly asks to analyze recent work and propose skills based on what they actually do.

BGPT Paper Search retrieves structured experimental data from scientific papers via the BGPT MCP server. Use it for literature reviews, evidence synthesis, or finding methods, results, sample sizes, and quality scores not available in abstracts alone.

BIDS helps organize, query, validate, and convert neuroscience datasets (MRI, EEG, MEG, PET, microscopy, behavioral) into the Brain Imaging Data Structure standard. Use it when preparing or sharing datasets for repositories like OpenNeuro and DANDI.

Benchling Integration enables programmatic access to Benchling's registry entities, inventory, notebook entries, workflows, and Data Warehouse through the Python SDK and REST API. Use it when automating life sciences R&D data or syncing Benchling with external systems.

BioServices provides a unified Python interface to 40+ bioinformatics web services for retrieving protein, pathway, compound, and sequence data. Use it when queries span multiple databases like UniProt, KEGG, ChEMBL, or Reactome and require consistent cross-database ID mapping.

Biopython provides tools for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access via Bio.Entrez. Use it for batch processing, custom bioinformatics pipelines, and BLAST automation.

Bulk RNA-seq takes raw FASTQ reads through QC, trimming, alignment, quantification, and differential expression, then hands off to pathway enrichment and publication figures. Use it when you need a reproducible, end-to-end bulk RNA-seq differential-expression workflow.

COBRApy performs constraint-based metabolic modeling for systems biology, including FBA, FVA, gene knockouts, flux sampling, and SBML model handling. Use it when working with genome-scale metabolic models to simulate and optimize cellular metabolism or design metabolic engineering experiments.

Cellxgene Census queries the CZ CELLxGENE Census programmatically for versioned single-cell and spatial transcriptomics data, including metadata, gene expression, summary counts, embeddings, and source H5AD files. Use it for large-scale bioinformatics analyses, machine learning on cell data, or cross-dataset comparisons.

Cirq designs, simulates, and runs quantum circuits on quantum computers and simulators. Use it for Google Quantum AI hardware, noise modeling, and quantum characterization experiments.

Citation Management finds, validates, and formats academic citations across OpenAlex, PubMed, and Google Scholar. Use it to convert DOIs/PMIDs to BibTeX, verify metadata, deduplicate references, and produce clean bibliographies for scientific writing.

Clinical Decision Support Research prepares and validates research-only clinical decision-support evaluation, evidence-profile, cohort, survival, biomarker/model, privacy, and governance artifacts. Use it for aggregate or synthetic research documentation and traceability—not patient care or live clinical operation.

Clinical Reports creates safety-bounded draft structures and runs local deterministic checks for clinical case, diagnostic, trial, safety, and aggregate research reports. Use it only with synthetic, de-identified, or aggregate inputs and verified source-fact manifests; every output requires qualified review.

Consciousness Council runs a structured multi-perspective deliberation, summoning 4-6 thinking archetypes to analyze a question and synthesizing their perspectives into actionable insight. Use it when the user wants diverse viewpoints, faces a tough decision, or requests a council, panel, or devil's advocate analysis.

Dask scales pandas and NumPy workflows to larger-than-memory datasets and across clusters. Use it when you need parallel file processing, distributed ML, or to scale existing pandas/NumPy code beyond memory.

Database Lookup queries documented public database APIs to retrieve reproducible facts from named sources. Use it when a scientific, regulatory, financial, or other database-backed fact must be retrieved reproducibly rather than inferred.

Datamol provides a Pythonic wrapper around RDKit for molecular cheminformatics, including SMILES parsing, standardization, descriptors, fingerprints, clustering, and 3D conformers. Use it for drug discovery and molecular data workflows.

DeepChem enables molecular property prediction, drug discovery, and materials design through molecular featurization, MoleculeNet benchmarks, and training of classical ML and graph neural network models. Use it for ADMET/toxicity prediction, transfer learning with pretrained chemistry models, or benchmarking on datasets like Tox21 and BBBP.