Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/danielrosehill/claude-data-analyst-plugin/multivariate-analysisnpx skills add danielrosehill/Claude-Data-Analyst-plugin --skill multivariate-analysisgit clone --depth 1 https://github.com/danielrosehill/Claude-Data-Analyst-pluginWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00070 | $0.01297 |
| Opus 5 | $0.00035 | $0.00648 |
| Sonnet 5 | $0.00014 | $0.00259 |
| Haiku 4.5 | $0.00007 | $0.00130 |
Grade A, and why
multivariate-analysis scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Multivariate Analysis
Go beyond pairwise correlation. Tell the user which variables actually drive the target once the others are accounted for, which variables are redundant, and which combinations reveal structure that no single pair shows.
When to use this vs. correlation-analysis
correlation-analysis— "is A related to B?" Pairwise ranking.multivariate-analysis— "with A, B, C, D all in play, which matter, which are redundant, and are there interactions or latent factors?"
If the user says "correlation" but lists more than two variables of interest, or has a target they want to "explain", this is the right skill.
Inputs
- Path to a dataset (CSV / Parquet / Excel / DuckDB table).
- Optional: target variable (the one to explain). If unset, do an unsupervised pass.
- Optional: candidate predictors. Default: all numeric columns + encoded low-cardinality categoricals.
- Optional: grouping / segment column (for stratified analysis).
Recommended CLI tooling
duckdb— correlation matrices, VIF precursors (regression residuals), standardisation.uv run --with pandas --with scipy --with statsmodels --with scikit-learn python -c '...'— partial correlation, VIF, PCA, factor analysis, OLS with interactions.
Procedure
Step 1 — Profile and prep
- Identify numeric columns. Encode low-cardinality categoricals (≤ ~20 levels) as one-hot or ordinal as appropriate.
- Report missingness per column. If >20% missing on any candidate, ask whether to drop the column, drop rows, or impute (median/mode or model-based).
- Standardise (z-score) numerics before anything scale-sensitive (PCA, regularised regression, distance-based methods).
Step 2 — Full correlation matrix
Compute the full pairwise correlation matrix (Pearson by default; switch to Spearman if any variable is heavily skewed or ordinal). Surface:
- The strongest pairs (top 10 by |r|).
- Clusters of mutually correlated variables (simple hierarchical clustering on
1 - |r|).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 104 lines · 70 tokens per session scan A 710f005d96a3
multivariate-analysis is a skill published in the GitHub repository danielrosehill/Claude-Data-Analyst-plugin (11 stars, last pushed 4mo ago), licensed MIT. It adds 70 tokens to every session and 1,297 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
addressing-pr-review-comments
Address all valid review comments on a PR for the current branch in the streamlit/streamlit repo. Covers both inline review comments and general PR (issue) comments. Use when a PR has reviewer feedback to address, including code changes, style fixes, and documentation updates.
fixing-streamlit-ci
Analyze and fix failed GitHub Actions CI jobs for the current branch/PR. Use when CI checks fail, PR checks show failures, or you need to diagnose lint/type/test errors and verify fixes locally.
improving-python-coverage
Runs Python unit tests with coverage, analyzes coverage reports, and implements meaningful tests to increase coverage by 0.2%. Use when you want to systematically improve Python test coverage with high-value test cases.
improving-frontend-coverage
Runs frontend unit tests with coverage, analyzes coverage reports, and implements meaningful tests to increase coverage by 0.2%. Use when you want to systematically improve frontend test coverage with high-value test cases.
reviewing-readability
Evaluates comments, docstrings, and naming in code for readability by a developer new to the codebase — is the documentation clear and concise, and is non-obvious logic documented? Produces findings with concrete proposed rewrites (or additions where documentation is missing); the caller decides whether to apply them…
reviewing-pr-description
Evaluates a PR's title and description for readability — do they clearly and concisely convey what changed and why to a reviewer? Produces findings with concrete proposed rewrites; the caller decides whether to apply them or present them as feedback. Use when finalizing a PR or reviewing PR metadata. For code…