OpenSkillIndex← back to search
claude-codescience

Babysitter bayesian-inference-engine

Bayesian probabilistic reasoning for prior specification, posterior computation, and belief updating

install
source · Clone the upstream repo
git clone https://github.com/a5c-ai/babysitter
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/a5c-ai/babysitter "$T" && mkdir -p ~/.claude/skills && cp -r "$T/library/specializations/domains/science/scientific-discovery/skills/bayesian-inference-engine" ~/.claude/skills/a5c-ai-babysitter-bayesian-inference-engine && rm -rf "$T"
manifest: library/specializations/domains/science/scientific-discovery/skills/bayesian-inference-engine/SKILL.md
tags
#bayesian-reasoning#probabilistic-modeling#mcmc-sampling#model-comparison#posterior-analysis
source content

Bayesian Inference Engine

Purpose

Provides Bayesian probabilistic reasoning capabilities for prior specification, posterior computation, and sequential belief updating.

Capabilities

  • Prior elicitation support
  • MCMC sampling (NUTS, HMC)
  • Variational inference
  • Model comparison (Bayes factors, LOO-CV)
  • Posterior predictive checking
  • Sequential belief updating

Usage Guidelines

  1. Prior Selection: Choose appropriate, defensible priors
  2. Sampling: Use efficient MCMC algorithms
  3. Diagnostics: Check convergence and mixing
  4. Model Comparison: Use appropriate comparison criteria

Tools/Libraries

  • PyMC
  • Stan (PyStan)
  • ArviZ
  • NumPyro