Menu
Skills neste repositório
O SkillsMP coletou 420 skills de eightmm/codex-science. Abra uma skill para revisar a origem e os detalhes.
eightmm/codex-scienceMostrando 40 de 420 skills coletadas.
Model interpretability and explainability using SHAP (SHapley Additive exPlanations). Use this skill when explaining machine learning model predictions, computing feature importance, generating SHAP plots (waterfall, beeswarm, bar, scatter, force, heatmap),…
Idioma do texto original: inglês
Statistical models library for Python. Use when you need specific model classes (OLS, GLM, mixed models, ARIMA) with detailed diagnostics, residuals, and inference. Best for econometrics, time series, rigorous inference with coefficient tables. For guided…
Idioma do texto original: inglês
Use when you need exact symbolic math in Python — algebra, calculus, equation solving, symbolic linear algebra, or code generation via lambdify/LaTeX. Prefer NumPy or SciPy when floating-point approximations are sufficient.
Idioma do texto original: inglês
Zero-shot time series forecasting with Google's TimesFM foundation model. Use for any univariate time series (sales, sensors, energy, vitals, weather) without training a custom model. Supports CSV/DataFrame/array inputs with point forecasts and prediction…
Idioma do texto original: inglês
Generate concise (3-4 page), focused medical treatment plans in LaTeX/PDF format for all clinical specialties. Supports general medical treatment, rehabilitation therapy, mental health care, chronic disease management, perioperative care, and pain management.…
Idioma do texto original: inglês
Access comprehensive LaTeX templates, formatting requirements, and submission guidelines for major scientific publication venues (Nature, Science, PLOS, IEEE, ACM), academic conferences (NeurIPS, ICML, CVPR, CHI), research posters, and grant proposals (NSF,…
Idioma do texto original: inglês
Chunked N-D arrays for cloud storage (Zarr-Python 3). Compressed arrays, parallel I/O, S3/GCS via fsspec, NumPy/Dask/Xarray compatible, for large-scale scientific computing pipelines.
Idioma do texto original: inglês
Run reproducible AlphaFold2 monomer or multimer inference with pinned code, parameters, sequence databases, template cutoff, and seeds. Use when a local AlphaFold2 installation or approved container/database setup is available; use AlphaFold DB lookup instead…
Idioma do texto original: inglês
Run the official AlphaFold3 inference pipeline for approved theoretical modeling of proteins, nucleic acids, ligands, modifications, and covalent complexes. Use only when the user has legitimately obtained model parameters and accepts the source, weight,…
Idioma do texto original: inglês
Construct and validate asymptotic and perturbative approximations. Use for dominant balance, regular or singular perturbations, boundary layers, matched expansions, multiple scales, WKB, Laplace methods, and stationary phase.
Idioma do texto original: inglês
Run reproducible local protein-ligand docking with AutoDock Vina and Meeko. Use for pose generation, redocking, or small virtual screens when a receptor and a justified binding pocket are available; do not use a Vina score as experimental affinity.
Idioma do texto original: inglês
Run a pinned BindCraft campaign for de novo protein binders using AlphaFold2 optimization, ProteinMPNN, PyRosetta, and configurable filters. Use only when the target structure, hotspot strategy, candidate budget, dependency licenses, and experimental handoff…
Idioma do texto original: inglês
Run pinned Borzoi models to predict RNA-seq and regulatory tracks from long DNA sequence and compare reference versus alternate alleles. Use for regulatory variant hypotheses with explicit genome assembly, sequence-window, tissue, and track provenance.
Idioma do texto original: inglês
Run pinned Chai-1 inference for proteins, complexes, nucleic acids, ligands, covalent bonds, templates, or restraints. Use when multimolecular structure prediction needs Chai-1-specific inputs and confidence outputs.
Idioma do texto original: inglês
Analyze nonlinear dynamical systems, bifurcations, and deterministic chaos. Use for flows and maps, phase portraits, fixed points, limit cycles, Lyapunov exponents, Poincare sections, attractors, continuation, and distinguishing chaos from noise or numerical…
Idioma do texto original: inglês
Integrate molecular formula, NMR, MS, IR, Raman, UV-Vis, chromatography, and diffraction evidence into ranked chemical structures with explicit contradictions and confidence. Use for unknown identification, structure confirmation, isomer discrimination,…
Idioma do texto original: inglês
Quantify analytes from chromatographic data with validated integration, calibration, quality controls, and uncertainty. Use for HPLC, UHPLC, GC, LC-MS, GC-MS, UV or fluorescence detection, targeted assays, purity estimates, retention data, and batch…
Idioma do texto original: inglês
Formulate, solve, simulate, and verify classical mechanics problems using Newtonian, Lagrangian, Hamiltonian, conservation-law, oscillation, rotation, and central-force methods. Use for particles, rigid bodies, constrained systems, collisions, orbital motion,…
Idioma do texto original: inglês
Solve and verify complex-variable and Fourier-analysis problems. Use for analytic functions, contour integrals, residues, branch cuts, Fourier series and transforms, convolution, spectral methods, sampling, and dispersion relations.
Idioma do texto original: inglês
Verify computational physics software, quantify numerical uncertainty, and validate models against experiments. Use for convergence studies, manufactured solutions, conservation audits, code-to-code comparisons, calibration, uncertainty quantification, and…
Idioma do texto original: inglês
Analyze condensed-matter and solid-state models and computations. Use for crystal symmetry, reciprocal space, electronic bands, density of states, Fermi surfaces, phonons, transport, magnetism, superconductivity, defects, and phase behavior.
Idioma do texto original: inglês
Formulate, solve, and verify continuum models for solids and fluids. Use for deformation, stress, strain, conservation laws, constitutive equations, elasticity, viscosity, boundary conditions, weak forms, and finite-element or finite-volume analysis.
Idioma do texto original: inglês
Model dynamical systems, analyze stability, and design or verify feedback controllers. Use for state-space and transfer-function models, controllability, observability, PID, state feedback, estimation, optimal control, robustness, discretization, and…
Idioma do texto original: inglês
Run a pinned DiffDock release for diffusion-based protein-ligand pose prediction. Use for candidate pose generation from receptor coordinates and ligand structures, especially without a fixed pocket; do not interpret DiffDock confidence as affinity.
Idioma do texto original: inglês
Formulate, solve, simulate, and verify electrostatics, magnetostatics, circuits, induction, electromagnetic waves, and boundary-value problems using Maxwell's equations, potentials, constitutive relations, and conservation laws.
Idioma do texto original: inglês
Extract reproducible ESM-2 protein embeddings, residue representations, likelihoods, or zero-shot mutation scores. Use when a frozen protein language model is needed as a feature extractor or baseline.
Idioma do texto original: inglês
Run released Biohub ESMC protein language models for representations, masked likelihoods, sequence scoring, mutation analysis, or sparse-autoencoder features. Use as the current ESMC alternative to ESM-2 when model scale, revision, and inference boundaries…
Idioma do texto original: inglês
Run the legacy public ESMFold v1 single-sequence structure model reproducibly for proteins or simple multimers. Use for the 2022 ESM-2-based model; use the separate ESMFold2 skill for Biohub's 2026 diffusion/all-atom model.
Idioma do texto original: inglês
Run Biohub ESMFold2 locally from the released Hugging Face weights or through the approved Biohub Platform. Use for all-atom prediction of proteins, complexes, DNA, modifications, and ligands with optional MSA input and diffusion sampling; distinct from the…
Idioma do texto original: inglês
Run pinned Evo 2 DNA sequence scoring, embeddings, variant scoring, or controlled generation. Use for long-context genomic foundation-model analyses when compatible local GPU hardware and explicit genome/strand provenance are available.
Idioma do texto original: inglês
Formulate and verify geometry and topology arguments. Use for metric and topological spaces, continuity, compactness, connectedness, manifolds, curves and surfaces, homotopy, covering spaces, Euler characteristic, and geometric invariants.
Idioma do texto original: inglês
Run reproducible local GNINA docking or CNN rescoring on protein-ligand systems. Use when GPU-assisted pose generation, refinement, or CNN reranking is wanted; keep CNN pose score, predicted affinity, and experimental affinity as distinct quantities.
Idioma do texto original: inglês
Prepare and run staged, checkpointed molecular dynamics with a pinned GROMACS build. Use when GROMACS workflows, HPC execution, established .mdp protocols, or interoperability with GROMACS topology/trajectory formats is required.
Idioma do texto original: inglês
Build an evidence-traceable indication dossier for a disease, target, mechanism, drug, biomarker, or therapeutic hypothesis. Use for landscape assessment, target-indication rationale, clinical pipeline review, translational gaps, and evidence-backed go/no-go…
Idioma do texto original: inglês
Formulate and solve inverse problems with stable regularization and calibrated uncertainty. Use for parameter or field recovery, deconvolution, tomography, system identification, data assimilation, Bayesian inversion, and learned inverse models.
Idioma do texto original: inglês
Solve exact and numerical linear algebra problems with explicit field, dimensions, bases, rank structure, conditioning, and residual checks. Use for linear systems, vector spaces, linear maps, eigenproblems, least squares, SVD, quadratic forms, and matrix…
Idioma do texto original: inglês
Identify and quantify support for analytes from mass spectrometry with explicit acquisition context, calibration, false-discovery control, and isomer-aware evidence. Use for GC-MS, LC-MS, MS/MS, accurate mass, isotope patterns, adducts, fragments, library…
Idioma do texto original: inglês
Analyze OpenMM, GROMACS, and other molecular dynamics trajectories with MDAnalysis using topology-aware selections, periodic-boundary handling, prespecified observables, convergence checks, and uncertainty estimates.
Idioma do texto original: inglês
Drive a concrete scientific modeling problem from supplied inputs through model selection, falsifiable plan, one-time approval, environment setup, smoke test, full execution, downstream analysis, provenance, and review. Use when the user provides sequences,…
Idioma do texto original: inglês
Prepare receptors and small molecules for docking or simulation with explicit stereochemistry, protonation, tautomer, charge, conformer, cofactor, water, and pocket provenance. Use before AutoDock Vina, GNINA, DiffDock, OpenMM, GROMACS, or any protein-ligand…
Idioma do texto original: inglês