By pymc-labs
Build and run reactive marimo Python notebooks for Bayesian modeling and EDA with PyMC v5+. Specify priors, run MCMC inference via nutpie, diagnose with ArviZ, and test models fast using pytest mocks without slow sampling. Trigger custom Python scripts on prompt submits.
npx claudepluginhub pymc-labs/python-analytics-skills --plugin analyticsALWAYS use when: creating/editing marimo notebooks, working with any .py file containing @app.cell decorators, building reactive Python notebooks, doing exploratory data analysis in notebook form, converting Jupyter (.ipynb) to marimo, or when user mentions "marimo", "reactive notebook", or asks for an interactive Python notebook. Covers marimo CLI (edit, run, convert, export), UI components (mo.ui.*), layout functions, SQL integration, caching, state management, and wigglystuff widgets. If a task involves notebooks and Python, invoke this skill first.
Bayesian statistical modeling with PyMC v5+. Use when building probabilistic models, specifying priors, running MCMC inference, diagnosing convergence, or comparing models. Covers PyMC, ArviZ, pymc-bart, pymc-extras, nutpie, and JAX/NumPyro backends. Triggers on tasks involving: Bayesian inference, posterior sampling, hierarchical/multilevel models, GLMs, time series, Gaussian processes, BART, mixture models, prior/posterior predictive checks, MCMC diagnostics, LOO-CV, WAIC, model comparison, or causal inference with do/observe.
Testing PyMC models with pytest. Use when writing unit tests for Bayesian models, setting up test fixtures, mocking MCMC sampling, or testing model structure. Covers pymc.testing.mock_sample, pytest fixtures, and the distinction between fast structure-only tests (mocking) and slow posterior inference tests. Triggers on: testing PyMC, pytest, unit tests for models, mock sampling, test fixtures, CI/CD for Bayesian models.
A plugin for Claude Code and other AI coding platforms providing Agent Skills for Bayesian modeling and reactive notebooks. Packages specialized knowledge for PyMC and marimo into skills that Claude loads on-demand.
| Skill | Description |
|---|---|
| pymc-modeling | Bayesian statistical modeling with PyMC v5+. Covers model specification, MCMC inference (nutpie, NumPyro), ArviZ diagnostics, hierarchical models, GLMs, GPs, BART, time series, and more. |
| pymc-testing | Testing PyMC models with pytest. Covers mock sampling with pymc.testing.mock_sample, pytest fixtures, and the distinction between fast structure-only tests (mocking) and slow posterior inference tests. |
| marimo-notebook | Reactive Python notebooks with marimo. Covers CLI, UI components, layout, SQL integration, caching, state management, and wigglystuff widgets. |
npx skills add pymc-labs/python-analytics-skills
One command, works with Claude Code, Cursor, Gemini CLI, and 15+ other agents.
Two-step process using Claude Code slash commands:
/plugin marketplace add pymc-labs/python-analytics-skills
/plugin install analytics@pymc-labs-python-analytics-skills
Installs all skills plus the keyword-suggestion hook. Supports /plugin update for future updates.
git clone https://github.com/pymc-labs/python-analytics-skills.git
cd python-analytics-skills
./install.sh claude # Claude Code
./install.sh all # All platforms
./install.sh claude -- pymc-modeling # Specific skill only
# List available skills with descriptions
./install.sh --list
# Validate skill structure
./install.sh --validate
| Platform | Install Location | Auto-Discovered |
|---|---|---|
| Claude Code | ~/.claude/skills/ | Yes |
| OpenCode | ~/.config/opencode/skills/ | Yes |
| Gemini CLI | ~/.gemini/skills/ | Yes |
| Cursor | ~/.cursor/skills/ | Yes |
| VS Code Copilot | ~/.copilot/skills/ | Yes |
python-analytics-skills/
├── .claude-plugin/
│ ├── marketplace.json # Plugin registry metadata
│ └── plugin.json # Plugin configuration
├── skills/
│ ├── pymc-modeling/
│ │ ├── SKILL.md # Main skill instructions
│ │ └── references/ # 12 detailed reference docs
│ ├── pymc-testing/
│ │ ├── SKILL.md # Main skill instructions
│ │ └── references/
│ └── marimo-notebook/
│ ├── SKILL.md # Main skill instructions
│ ├── references/ # 4 reference docs
│ ├── assets/ # Notebook templates
│ └── scripts/ # Conversion utilities
├── hooks/
│ ├── hooks.json # Hook configuration
│ └── suggest-skill.sh # Keyword-based skill suggestion
├── install.sh # Multi-platform installer
├── package.json # npm package metadata
└── skills.json # Skills registry
The plugin includes a UserPromptSubmit hook that suggests relevant skills when it detects keywords in your prompt:
Skill not loading:
SKILL.md./install.sh --validate to check structureclaude --debug for hook/skill loading errorsHook not firing:
/hooks in Claude Code to see loaded hooksecho '{"user_prompt": "bayesian model"}' | bash hooks/suggest-skill.shSee CONTRIBUTING.md for guidelines on adding new skills.
MIT License. See LICENSE for details.
marimo-team/skills bundled as a Claude Code plugin
Share bugs, ideas, or general feedback.
Comprehensive model evaluation with multiple metrics
Evaluate and compare ML model performance metrics
Scientific research agent extension - turns research goals into reproducible Jupyter notebooks with Python REPL, data analysis, and ML workflows
Editorial "Business Analyst" bundle for Claude Code from Antigravity Awesome Skills.
Multi-agent workflow framework for building, testing, and shipping statistical software packages