Pymc

This gives Claude the ability to build and validate Bayesian models using PyMC's modern API. It covers the full workflow: prior and posterior predictive checks, MCMC sampling with NUTS, model diagnostics (R-hat, ESS, divergences), and model comparison via LOO and WAIC. The included templates handle common patterns like hierarchical models with non-centered parameterization, which is critical for avoiding sampling issues. Use this when you need proper uncertainty quantification, want to work with hierarchical data structures, or need to handle missing data in a principled way. The diagnostic scripts are helpful since convergence checking is non-negotiable in Bayesian workflows. It assumes you understand why you're reaching for Bayesian methods rather than maximum likelihood approaches.

skills.sh installs: 2.1K
GitHub stars: 56.6K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add k-dense-ai/scientific-agent-skills --skill pymc --agent claude-code

Installs into .claude/skills of the current project.

Files

PyMC Bayesian Modeling

Version and scope

Targets PyMC 6.3.2, PyTensor 3.3.2, and ArviZ 1.3.0. These are local Python APIs, with no service endpoint, credentials, or remote inference required. The native tests use the PyMC NUTS sampler on small synthetic models. Optional nutpie/JAX samplers, SMC, and production-scale convergence are not validated by those tests. Record the complete resolved environment, backend, seeds, preprocessing, and data provenance for a scientific run.

Local macOS validation used PYTENSOR_FLAGS="cxx=" after the C linker failed with library 'd64' not found; this disables PyTensor C compilation for that process. The Python library workflows passed under that setting; it does not establish that the C backend works on this host.

SKILL.md on GitHub

Details

Repository: github.com/k-dense-ai/scientific-agent-skills
Categories: DevOps & CI/CD, AI & Agent Building
First seen: Jun 3, 2026