ai-analyst-lab/ai-analyst-plugin
Plugin Claude Code
The full AI Analyst skill system for data analysis in Claude Cowork: decision framing, data profiling, validated analysis, visualization, and a working-folder memory that learns your corrections.
ai-analyst-lab/ai-analyst-plugin
Agent
Execute the selected causal inference method on data using the causal skill's bundled causalstats package. Produces point estimates, confidence intervals, and diagnostic charts.
ai-analyst-lab/ai-analyst-plugin
Agent
Per-method diagnostic battery that checks whether the assumptions required for the selected causal method hold in the data.
ai-analyst-lab/ai-analyst-plugin
Agent
Interactive decision tree for selecting the right causal inference method based on data availability, study design, and assumptions.
ai-analyst-lab/ai-analyst-plugin
Agent
Generate a complete causal inference report with mandatory sections including method choice, results, assumption checks, sensitivity analysis, confidence assessment, and non-negotiable caveats.
ai-analyst-lab/ai-analyst-plugin
Agent
Sensitivity analysis for observational causal estimates — Rosenbaum bounds, E-values, and placebo tests to assess robustness to unmeasured confounding.
ai-analyst-lab/ai-analyst-plugin
Agent
Adversarial agent that finds threats to analytical validity — concurrent changes, data quality issues, selection biases, and measurement artifacts that could invalidate an analysis.
ai-analyst-lab/ai-analyst-plugin
Agent
Full experiment analysis workflow — from data loading through SRM validation, treatment effects, segment analysis, novelty checks, guardrail evaluation, and nuanced ship/kill/iterate recommendation.
ai-analyst-lab/ai-analyst-plugin
Agent
Design experiments or quasi-experimental analyses to test causal hypotheses, including power estimation, guardrail selection, and pre-registered decision rules.
ai-analyst-lab/ai-analyst-plugin
Agent
Walk the Result Interpretation Tree to classify experiment outcomes as Ship/Abort/Learn/Invalid using Spotify's EwL framework. References pre-registered decision rules from experiment.yaml.
ai-analyst-lab/ai-analyst-plugin
Agent
Daily monitoring for running experiments — SRM trending, guardrail status, sample accumulation tracking, and power projection. Uses coded helpers from experimentstats.
ai-analyst-lab/ai-analyst-plugin
Agent
Transform experiment analysis results into a stakeholder-ready readout with executive summary, visualizations, per-segment decisions, ramp plan, and follow-up experiment proposals.
ai-analyst-lab/ai-analyst-plugin
Agent
Takes V1 analysis findings and messy stakeholder feedback (comments, meeting transcripts, Slack threads), categorizes the feedback, identifies what V1 got right vs. wrong, and produces a structured V2 investigation plan with a stakeholder answer map.
ai-analyst-lab/ai-analyst-plugin
Agent
Takes a vague analytical hunch and transforms it into a testable hypothesis with precise metrics, comparison groups, key segments, natural experiments, and accept/reject criteria.
ai-analyst-lab/ai-analyst-plugin
Command
Start an analysis the AI Analyst Plus way, framing the decision first and running the method end to end.
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Never present a metric or number in isolation; anchor every number to a comparison (prior period, benchmark, or another segment) or state that none is available. Use every time you show a number: rates, counts, revenue, averages, query results, chart labels, summary stats. Trigger on output containing "the rate is"…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Turn a vague hunch or stakeholder request into a rigorous analysis plan via a staged pipeline (hypothesis sharpening, confound scan, V1, feedback, V2 redesign). Trigger on "I think X caused Y", "design an analysis for", "what's driving", "V2 after feedback", "stakeholder comments", "redesign the analysis". Vs…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Operating rules for every data analysis. Apply for ANY data-analysis intent: "analyze", "investigate", "why did X change", "compare", "report on", "dashboard", "metrics", "funnel", "retention", "revenue", "conversion", "trend", "segment", "forecast", "how are we doing", "dig into", "break down", or any question about…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Retrieve proven SQL patterns, table cheatsheets, and join patterns from .knowledge/query-archaeology/ so past work gets reused. Fire as a pre-flight step before writing ANY SQL. Also trigger on "do we have a known query for X", "how do we usually join these tables", "have we computed this metric before". If the store…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Browse and search the organization's business context in .knowledge/organizations/: glossary terms, product catalog, metric definitions, OKRs, team structure. Trigger on "/business", "show me our glossary", "what metrics are defined?", "what products do we have?", "show me our OKRs", "who's on which team?", "look up…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Causal inference toolkit for when experiments are not possible: estimate treatment effects from observational data with assumption checks and mandatory caveats. Invoke as /causal. Trigger on "causal", "caused", "impact of", "effect of", "attribution", "counterfactual", "difference-in-differences", "DiD", "propensity…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
End every analysis that makes a recommendation with an actionable follow-up plan. Use ONLY AFTER a specific recommendation ("we should...", "I recommend..."); add the checklist automatically even if not asked. DO NOT use for pre-recommendation exploratory questions like "should we investigate this?" or "what's causing…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Every chart ships with 1-2 sentences of commentary, baked into the chart subtitle AND a sidecar .md, telling the reader what to NOTICE, not what the chart depicts. Trigger every time you save, render, embed, or caption any visual, and when presenting or re-sharing a chart that ALREADY EXISTS: regenerate it if its…
ai-analyst-lab/ai-analyst-plugin
Skill Claude CodeCodex
Compare metrics, findings, and patterns across two or more connected datasets. Trigger on "compare datasets", "is this pattern the same across", "how does X differ between datasets", "cross-dataset comparison", "compare metrics between", or asking whether a finding is unique to one dataset. Also offer proactively…