pymc-labs
- 56 skills
- 0 followers
- 3 hours ago last updated
- ▌ Author Skills · pymc-labs bundleAuthor, configure, and distribute Agent Skills for a Great Docs site. Covers the three scenarios: automatic skill generation, adding a single hand-written skill, and distributing multiple named skills with a switcher page. Use when creating, editing, or configuring SKILL.md files for package documentation.
- ▌ Configure Site · pymc-labs bundleConfigure a Great Docs documentation site through great-docs.yml. Covers theming (navbar gradients, content glow, dark mode), hero sections, logos, announcements, sidebar options, page tags, page status badges, SEO, analytics, and deployment settings. Use when customizing site appearance, enabling features, or tuning build behavior.
- ▌ Write User Guide · pymc-labs bundleWrite and maintain narrative user-guide pages for a Great Docs site. Covers page creation, QMD frontmatter, section grouping, sidebar ordering, callouts, executable code cells, cross-references, and content guidelines. Use when adding, reorganizing, or improving user-guide content.
- ▌ Revise Docstrings · pymc-labs bundleReview and improve Python docstrings for Great Docs API reference generation. Covers NumPy and Google style conventions, parameter documentation, return types, examples, cross-references, and Great Docs directives (%seealso, %nodoc). Use when auditing, writing, or fixing docstrings in a Python package.
- ▌ Tdd · pymc-labs bundleTest-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
- ▌ Code Review · pymc-labsReview the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/spec asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".
- ▌
- ▌ Research 2 · pymc-labsStructure a research based on the user request. Identify what must change in order to complete the task.
- ▌ Make Plan 2 · pymc-labsCreate a plan based on document research through an interactive, iterative process.
- ▌ Dhub CLI · pymc-labs bundleGuide for using the dhub CLI — the AI skill manager for data science agents. Covers authentication, publishing, installing, running skills, managing API keys, eval reports, and troubleshooting. Use when users ask about dhub commands, skill publishing workflows, or need help with the Decision Hub CLI.
- ▌ Dhub Skill Creator · pymc-labs bundleGuide for creating effective skills for Claude Code agents. Covers skill design, implementation, validation, packaging, and optionally runtime environments and automated evaluations for Decision Hub publishing. Use when users want to create, improve, or package a skill.
- ▌ Pymc Extras · pymc-labs bundleLoad when the user is working with pymc-extras (pmx) features: splines / BSplineBasis, distributional regression / GAMLSS, R2D2M2CP or horseshoe priors, discrete variable marginalization, or Laplace approximation via fit_laplace. Triggers include: pymc_extras, pymc-extras, pmx, splines, BSplineBasis, distributional regression, GAMLSS, R2D2, horseshoe (regularized/Finnish), marginalize, fit_laplace, penalized splines.
- ▌ Pymc Testing · pymc-labs bundleLoad when writing or modifying pytest tests that touch pymc.Model, pm.sample, or any PyMC model code. Covers pymc.testing.mock_sample, pytest fixtures for Bayesian models, and the distinction between fast structure-only tests (mocking) and slow posterior inference tests. Triggers include: testing PyMC, pytest with pymc, unit tests for Bayesian models, mock sampling, test fixtures for models, CI/CD for PyMC.
- ▌ Model Evaluation · pymc-labs bundleLoad when the user is comparing Bayesian models, computing LOO-CV / ELPD, calling az.loo or az.compare, doing model stacking/averaging, or computing Bayes factors. Covers the ArviZ 1.0 LOO/ELPD/stacking APIs exclusively (no waic). Triggers include: model comparison, LOO, ELPD, az.compare, az.loo, loo_expectations, loo_metrics, loo_r2, Pareto k, stacking, Bayes factor, cross-validation, predictive accuracy, information criterion.
- ▌ Prior Elicitation · pymc-labs bundleLoad when the user is choosing priors, running prior predictive checks, calling find_constrained_prior, using PreliZ, or otherwise eliciting domain knowledge into a Bayesian model. Covers weakly informative priors, constrained priors, sensitivity analysis, and elicitation workflows. Triggers include: prior selection, elicitation, find_constrained_prior, PreliZ, prior predictive, expert/informative priors, weakly informative priors, constrained priors.
- ▌ Work · pymc-labsState-aware orchestrator for enhancements and features. Detects where work left off (spec, implement, review) and advances it one phase. Delegates to grill-with-docs, to-spec, implement, tdd, and code-review internally — the user invokes only this skill. Use when the user says "work #N", "work on #N", "advance #N", or "work" to see in-flight items.
- ▌ Pathmc · pymc-labsBayesian path analysis (observed-variable SEM) in PyMC. Compiles a lavaan-inspired formula DSL into a generative PyMC model, then layers introspection, identification diagnostics, the `do()` operator, and causal estimands (ATE/CATE/ATT/ATU/prob) on top. Use when the user asks to specify, fit, or query a Bayesian structural causal model; estimate average treatment effects via g-computation; check identification with adjustment sets or the front-door criterion; or simulate panel/longitudinal counterfactuals.
- ▌ Fix Bug · pymc-labsAutonomous bug-fix workflow. Guides the orchestrator through state detection, implementation, test/lint validation, and a bounded fix/review loop with an independent reviewer subagent. Use when fixing a GitHub bug report, or when the user says "fix bug", "bugfix", or references a bug issue or PR such as "fix bug #149", "bugfix #149", or "fix bug PR #123".
- ▌ Great Docs · pymc-labs bundleGenerate documentation sites for Python packages with Great Docs. Covers init, build, preview, configuration (great-docs.yml), API reference, CLI docs, user guides, theming, deployment, and the llms.txt agent-context files. Use when creating, configuring, building, or troubleshooting Python package documentation.
- ▌ Mmm Modeling · pymc-labs bundleMedia Mix Modeling with PyMC-Marketing. Use when building MMMs, specifying adstock/saturation transformations, setting priors, fitting multidimensional (geo-level) models, computing channel contributions, ROAS, running budget optimization, calibrating with lift tests, or performing sensitivity analysis. Covers the MMM class, GeometricAdstock, LogisticSaturation, BudgetOptimizerWrapper, and ArviZ diagnostics for marketing models.
- ▌
- ▌ Research · pymc-labsStructure a research based on the user request. Identify what must change in order to complete the task.
- ▌
- ▌ Make Plan · pymc-labsCreate a plan based on document research through an interactive, iterative process.
- ▌
- ▌ Code Best Practice · pymc-labsPyMC-Marketing coding conventions, preferred implementations, and style guidelines.
- ▌ Working With Notebooks · pymc-labsFixing and validating Jupyter notebooks in pymc-marketing. Use when notebooks fail CI, need arviz_plots 2.x migration fixes, or when source code changes require notebook output updates.
- ▌
- ▌ Daimon Context · pymc-labsUse when the user asks about team context, project status, decisions, or discussions that live in Slack or Discord, or names a daimon. Routes the question to the right daimon servers and merges their answers.
- ▌ Data Cleaning · pymc-labsMutate data without destroying the evidence — dedup, dtype coercion, missing-value decisions, outlier flagging, category and unit normalisation, joins, pivots and aggregations to a new grain, and columns the source does not contain — appending a row to run/changelog.jsonl for every operation, whether what you write is one parquet file, one part per table, parquet parts of a log too large to load, a saved warehouse query, or cleaned files plus their index frame. The source is never edited in place, and under pandas 3 copy-on-write a chained assignment like df[mask]["col"] = 0 silently changes nothing. Use when a validation check failed, when the grain has duplicates, wrong dtypes or unparseable dates, when a source is too large or raw to query repeatedly and must be materialised once as typed parquet parts, when a collection needs an index frame derived, when a metric or label no column holds must be defined, or when asked to clean, dedup, impute, join, reshape or fix a dataset.
- ▌ File Handling · pymc-labsRead images and large files without destroying the conversation. Use before reading any image — a screenshot you took, a chart you rendered, a photo the user sent — and before reading a file you have not sized. Reading an oversized image ends the conversation permanently and cannot be undone.
- ▌ Report Reader · pymc-labsAnswer questions about a published report from the analysis bundle mounted in this session. Covers unpacking the bundle, tracing a number back to its source table, model or raw data, refitting a model on request, and rebuilding and re-uploading the report. Use whenever you are a reader-facing agent for one published report.
- ▌ Data Ingestion · pymc-labsLoad whatever the request points at — one CSV or Excel export, twelve related tables, a folder of 500 emails, a 4 GB log you never load, a warehouse table you query in place — and record every source in run/manifest.json with the grain one row represents, the axis it is ordered by, its row count, columns and dtypes. Read as text and coerce on purpose — pd.read_csv inference turns order_id 00123 into the integer 123, and a paged pull that stops early looks identical to a complete one. Use when a user attaches a file or a folder, points you at a database, warehouse or API, or asks for analysis of data you have not loaded yet.
- ▌ Report Publish · pymc-labsPublish a finished report as a page a named set of people can read and ask questions about. Covers gathering recipients and a cap, building the one archive the reading room expects, the size discipline for a large bundle, and the mint-then-upload-then-share procedure. Use when someone wants to hand a finished report to specific people, not for sharing a notebook or a chart in the thread.
- ▌ Data Validation · pymc-labsDecide whether data is fit to analyze before anyone analyzes it. A phase-1 pass runs in seconds — dtypes, the grain one row claims to represent, referential gaps between tables, plausibility ranges, coverage — and emits run/validation.json with a pass/warn/fail verdict; a fail is a stop, not a to-do item. Works the same on a flat export, twelve related tables, a keyless sensor stream, an index frame over 500 documents, or a warehouse table you check with SQL and never load. Phase 2 hunts the structural break — a definition, unit, currency or timezone that changed partway along whatever the data is ordered by. Use when data arrives from ingestion, before any chart, join or model, or when a number looks wrong and you cannot say why.
- ▌ Workspace Setup · pymc-labsFirst-time workspace setup and agent-roster operations for a Daimon workspace — working repo, keys, skills, MCP servers, routines, and which changes need an admin.
- ▌ Eda Storytelling · pymc-labsTurn a finished analysis into something a person reads and acts on — lead with the finding that settles the question or moves the decision, trace every number back to a run/findings.jsonl, run/changelog.jsonl or run/manifest.json record, and cut the rest. Works from whatever the analysis produced — one flat export, twelve joined tables, a warehouse table queried in place, an index frame over a corpus, a fitted model's posterior. Carries the data-validation verdict, the data-cleaning change log and any stage nobody ran into the writeup as caveats instead of burying them. Use when a single data question has to be answered in one chat reply, when writing up an analysis, when drafting a report, summary or readout of what the data or a fitted model showed, or when a pile of charts and tables needs an argument.
- ▌ Marimo Notebooks · pymc-labsPublish interactive marimo notebooks via the daimon MCP server. Mint a one-time upload URL with create_notebook_upload_url, get the .py into a sandbox file, and curl -X PUT --data-binary it to the URL — source never goes through a tool argument, which truncates. permanent=True publishes the same notebook as a read-only shareable blog instead of a scratch one. Also covers attaching data files, list_notebooks and delete_notebook. Use when someone asks for a notebook, dashboard, data explorer, or to publish an analysis as a blog post.
- ▌ Pymc Artifact Style · pymc-labs bundleApply PyMC Labs' house style to every artifact you produce — reports, PDFs, slide decks, charts, images, notebooks. Use whenever you generate something a person will look at, before you deliver it.
- ▌ Exploratory Data Analysis · pymc-labsProfile and interrogate data before anyone makes a claim about it — a cleaned frame, a folder of parquet parts, a warehouse table you query in place, or an index frame over a corpus of documents. Summary statistics do not identify a distribution — Anscombe's quartet shares a mean, variance and correlation across four unrelated shapes — so nothing is reported that has not been plotted. Covers dtype and cardinality profiling, missingness structure, Spearman against Pearson, a correlation matrix read as blocks instead of skimmed for its biggest cells, subgroup checks and whether one named group, batch or run is an outlier against the rest, variation along whatever index orders the data, writing one row per finding to run/findings.jsonl and handing it on to modeling. Use when asked to explore or profile a dataset, before a model is fit, when one group, device, batch or period looks wrong and you need to say how it differs, or when a pile of numbers has to become defensible findings.
- ▌ Dlab CLI · pymc-labs bundleComplete reference for decision-lab (dlab). Use when the user asks about creating decision-packs, designing data science agents, running sessions, analyzing results, or anything related to dlab CLI, agent architecture, parallel subagents, or decision-pack configuration. Covers the full workflow from scaffolding to analysis.
- ▌ Dlab Figure Style · pymc-labs bundledecision-lab house figure style for matplotlib. Use whenever creating, styling, or saving any matplotlib figure, chart, or plot. The environment is already styled — this skill covers only the rules the style config cannot enforce.
- ▌
- ▌
- ▌ Create Decision Pack Programmatically · pymc-labsHow to create a dlab decision-pack directory using generate_dpack() from Python code
- ▌ Analyze Dlab Session Runs · pymc-labsNavigate and analyze completed dlab session directories. Use when pointed at a work directory to understand what happened during a run — explore logs, outputs, parallel agent results, and the skills/prompts that shaped the analysis.
- ▌ Design Data Science Agent Systems · pymc-labsDesign agent system prompts, parallel architectures, and methodological guardrails for data science decision-packs. Use when creating orchestrator, subagent, or parallel agent systems for analytical workflows. Covers anti-fabrication rules, epistemic humility, when to stop, conflict detection, uncertainty reporting, retry protocols, prompt design principles, and the decision-lab runtime mechanics.
- ▌ Create Decision Pack Interactively · pymc-labsGuide a human through creating a dlab decision-pack by asking questions and then calling generate_dpack(). Use this skill whenever the user wants to create, set up, or scaffold a new decision-pack, agent environment, or Docker-sandboxed config for dlab — even if they don't use the word "decision-pack" explicitly. Trigger on phrases like "set up a new agent", "create an environment for X", "I want to run opencode for Y", "scaffold a project", or "make a new config".
- ▌ Pymc Data Handling · pymc-labsExpert on PyMC data management including pm.Data and pm.Minibatch for handling datasets, updating data containers, and mini-batch training. Use for data container errors or dataset handling issues.
- ▌ Pymc Samplers · pymc-labsExpert on PyMC MCMC sampling methods including NUTS, HMC, Metropolis variants, and pm.sample() API. Use for sampling errors, convergence issues, sampler configuration, or trace-related problems.
- ▌ Pymc Distributions · pymc-labsExpert on PyMC probability distributions including continuous (Normal, Beta, Gamma), discrete (Poisson, Binomial), multivariate (MvNormal, Dirichlet), mixture, and timeseries distributions. Use when encountering distribution errors, parameter issues, or migrating PyMC3 distribution code.
- ▌ Informative Priors For Mmm · pymc-labsExpert guide on calculating and setting informative priors for PyMC-Marketing MMM models based on data characteristics and domain knowledge. Use when configuring priors for intercept, channel effects, or adstock parameters.
- ▌ Pymc Marketing Mmm · pymc-labsExpert on PyMC-Marketing's Marketing Mix Model (MMM) framework including adstock transformations, saturation functions, hierarchical models, and GAM components. Use for MMM modeling, prior configuration, or pymc-marketing API questions.
- ▌ Event Forecasting · pymc-labs bundleMethodology for probabilistic forecasting of when and whether a future event will occur. Covers Bayesian survival models, reference class reasoning, driver threshold models, leading indicator models, scenario decomposition, and causal mechanism models. Use for any question of the form "When will X happen?" or "What is the probability that Y occurs by date Z?"
- ▌ Skill Name · pymc-labs bundleA clear description of what this skill does and when to use it. Include specific trigger keywords and task types that should activate this skill. Be specific about capabilities so the AI can decide when to load it. Maximum 1024 characters.
- ▌ Pymc Modeling · pymc-labs bundleBayesian 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.