CLI Tools for Efficient MCMC Inference in TuringLang

This project develops agent-accessible CLI tooling for the Julia Bayesian ecosystem, delivering three main components: DoodlePPL CLI for model validation, code generation, sampling, and diagnostics with structured JSON output, MCMCChains.js, a TypeScript npm library enabling MCMC diagnostics and visualizations in the browser and TuringCLI, a command-line interface for Turing.jl that provides machine-readable JSON outputs, with all three deliverables building upon and extending the GSoC 2025 work on DoodleBUGS.