Back to authors
sunxd3

sunxd3

16 Skills published on GitHub.

artifact-guidelines

Style and format conventions for subagent artifacts — reports, logs, figures.

UncategorizedView skill →

analysis-design

Methodology for setting up a Bayesian analysis — analysis purpose, validation strategy, domain context, structural questions. Read before specifying any models.

UncategorizedView skill →

bayesian-model-diagnostics

Interpreting LOO diagnostics, PIT calibration, and Pareto k for single-model evaluation

UncategorizedView skill →

bayesian-model-selection

Comparing model populations via ELPD, decision rules for selection vs stacking, and metric validity

UncategorizedView skill →

convergence-diagnostics

MCMC convergence diagnostics — thresholds, visual checks, and interpretation of HMC pathologies.

UncategorizedView skill →

eda

Exploratory data analysis reference for Bayesian modeling — operational EDA procedures (data semantics audit, quality checks, timestamp handling, visualization, modeling handoff) plus a diagnostic test library indexed by data shape.

UncategorizedView skill →

fake-data-simulation

Methodology for fake-data simulation — drawing or fixing true parameter values, simulating data from the model, refitting, and checking that inference recovers the parameters. Covers single-draw checks (cheap) and Simulation-Based Calibration (rigorous).

UncategorizedView skill →

generative-model-design

Required decisions for complete generative model specifications, the experiment-design discipline behind a resolution sequence, and the cross-cutting modeling principles that apply to every spec.

UncategorizedView skill →

inferencedata-handling

ArviZ InferenceData creation from CmdStanPy — variable naming conventions, conversion arguments, common failures.

UncategorizedView skill →

model-critique

Three-part critique methodology for a single fitted Bayesian model — statistical, domain, and framework assessment plus the verdict structure and refinement-suggestion conventions.

UncategorizedView skill →

orchestration

End-to-end Bayesian modeling workflow orchestration — phases (EDA → design → development → reporting), task-pool semantics for the validation pipeline, canonical file structure, and dispatch protocol for the eleven subagents. Loaded by the `/bayesian-workflow:run` command.

UncategorizedView skill →

python-environment

Reference for the Bayesian workflow Python environment — `shared_utils` API and script structure conventions. Assumes the environment has been bootstrapped via `/bayesian-workflow:setup`.

UncategorizedView skill →

report-writing

The writing quality bar for the final report — skim test, narrative arc, number density, figure discipline, detail tiers, and mechanical hygiene. Loaded by the report-writer agent; audit your draft against these rules before delivering.

UncategorizedView skill →

stan

Best practices for writing efficient, clean Stan programs. Covers structure, parameterization, prior/posterior predictive blocks, and pitfalls. Sub-references handle specialized cases.

UncategorizedView skill →

validation-protocol

Standard entry/exit protocol for pipeline subagents — input validation (arguments, filesystem, strict single-line [EXCEPTION] output on failure), the completed-work short-circuit, and the status.json completion record.

UncategorizedView skill →

visual-predictive-checks

Guidelines for visual predictive checks following Säilynoja et al. recommendations using ArviZ

UncategorizedView skill →