pymc
Bayesian modeling with PyMC. Build hierarchical models, MCMC (NUTS), variational inference, LOO/WAIC comparison, posterior checks, for probabilistic programming and inference.
85
83%
Does it follow best practices?
Impact
88%
1.49xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
82%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description with excellent specificity and domain-appropriate trigger terms that clearly carve out a distinct niche in Bayesian probabilistic programming with PyMC. Its main weakness is the lack of an explicit 'Use when...' clause, which would help Claude know precisely when to select this skill over others. Adding trigger guidance would elevate this from good to excellent.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user asks about Bayesian analysis, probabilistic programming, PyMC models, MCMC sampling, or statistical inference with uncertainty quantification.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Build hierarchical models', 'MCMC (NUTS)', 'variational inference', 'LOO/WAIC comparison', 'posterior checks'. These are concrete, well-defined capabilities in the Bayesian modeling domain. | 3 / 3 |
Completeness | The 'what' is well-covered with specific capabilities, but there is no explicit 'Use when...' clause or equivalent trigger guidance. The description only implies when it should be used through the listed terms. Per rubric guidelines, a missing 'Use when...' clause caps completeness at 2. | 2 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'Bayesian modeling', 'PyMC', 'hierarchical models', 'MCMC', 'NUTS', 'variational inference', 'LOO', 'WAIC', 'posterior checks', 'probabilistic programming', 'inference'. These cover the key terms a user working in this domain would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: Bayesian/probabilistic modeling specifically with PyMC. The specialized terminology (NUTS, LOO/WAIC, hierarchical models) makes it very unlikely to conflict with other skills like general statistics or machine learning skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, highly actionable skill with a clear multi-step workflow, explicit validation checkpoints, and good progressive disclosure to supporting files. Its main weakness is verbosity — the inline distribution guide, multiple model patterns, and troubleshooting sections make it quite long, and some of this content could be more concise or delegated to reference files. Overall it's a strong skill that effectively teaches Bayesian modeling with PyMC.
Suggestions
Move the Distribution Selection Guide and some Common Model Patterns to reference files to reduce the main skill's token footprint, keeping only the most essential pattern (e.g., hierarchical with non-centered parameterization) inline.
Trim explanatory text that Claude already knows, such as 'PyMC is a Python library for Bayesian modeling and probabilistic programming' and distribution category descriptions like 'For count data' or 'For binary outcomes'.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is quite long (~400 lines) and includes some content Claude would already know (e.g., basic distribution descriptions, what LOO/WAIC are). The distribution selection guide and common model patterns sections, while useful, could be more concise or offloaded to reference files. However, most content is practical and not purely explanatory. | 2 / 3 |
Actionability | The skill provides fully executable, copy-paste ready code examples throughout — model building, sampling, diagnostics, predictions, and multiple model patterns. Script invocations are concrete with specific function names and parameters. | 3 / 3 |
Workflow Clarity | The 8-step standard Bayesian workflow is clearly sequenced with explicit validation checkpoints (prior predictive check before fitting, diagnostics check before interpretation, posterior predictive check for validation). Each step includes specific checks and remediation steps for failures (e.g., divergences → increase target_accept, use non-centered parameterization). | 3 / 3 |
Progressive Disclosure | The skill provides a clear overview with well-signaled one-level-deep references to `references/distributions.md`, `references/sampling_inference.md`, `references/workflows.md`, `scripts/model_diagnostics.py`, `scripts/model_comparison.py`, and template files in `assets/`. The Resources section clearly describes when to use each reference file. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (571 lines); consider splitting into references/ and linking | Warning |
metadata_version | 'metadata.version' is missing | Warning |
Total | 9 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
25e1c0f
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.