Everyday Causal Skills plugin
Step-by-step causal inference: 14 skills pick a method, check assumptions, write R or Python code and stress-test the result.
- 84 ★ on the repository
- Last commit
Extended access
Why this level
- The skills write and run analysis code in your R or Python environment
- Power calculations and estimates rely on your data, which the agent reads from disk
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/everyday-causal-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skills write and run analysis code in your R or Python environment
- Power calculations and estimates rely on your data, which the agent reads from disk
This is third-party code. Review the repository files before installing.
What it does
The plugin follows the question "did it work or did we get lucky" from wording to report. `/causal-planner` takes a problem in plain language and recommends a method with an analysis plan. The method skills follow: A/B tests with power calculation, difference-in-differences, instrumental variables, regression discontinuity, synthetic control, matching, heterogeneous effects and interrupted time series. Each runs five stages: setup, assumptions, implementation, robustness, interpretation. `/causal-dag` helps draw a causal graph and find adjustment sets. `/causal-auditor` checks a finished analysis against five groups of validity threats. `/causal-roi` turns an effect estimate into money, and `/causal-report` compiles a final document in business, academic or hybrid mode. Results are not interpreted until the agent has seen actual code output, not just the code.
Who it is for. Marketing, product and BI analysts who must prove the effect of a launch instead of settling for a before-and-after comparison.
Good fit when
- An online store redesigned checkout and you must size an A/B test before launch
- A loyalty program went live in 12 of 50 stores and you need to know whether repeat purchases grew because of it
- The method is unclear: `/causal-planner` picks a suitable one from your data description and explains why
- A finished analysis should run through a list of validity threats before it goes to leadership
- The effect is estimated and has to become an ROI figure with a breakeven point
Not a fit when
- You need plain descriptive statistics or a forecast: the skills answer questions about cause
- You need accounting ROI, stock returns or real estate math: `/causal-roi` does not trigger on such requests
- You need a final verdict without specialist review: the author warns that AI can err on context-dependent assumptions
Example request
`/causal-planner` We launched a loyalty program in 12 of our 50 stores and the rest did not get it. Did repeat purchases grow because of the program?Limitations
The skills write R or Python code, so the matching packages must be present. If a package is missing or a plot fails, the report includes the code and offers help with environment setup. `/causal-roi` needs a ready effect estimate from other skills or from you and does not compute the effect itself. `/causal-dag` records your assumptions, not facts: a correct graph depends on your domain knowledge. The author notes that the plugin does not replace the book Everyday Causal Inference or an analyst judgment.
How to disable. In Claude Code remove the plugin with `/plugin uninstall everyday-causal-skills@everyday-causal-skills`. For Codex delete the copied `causal-*` folders from `~/.agents/skills/`; for Cursor delete the clone from `~/.cursor/plugins/local/`.
FAQ
How do I install the plugin in Claude Code?
Run `/plugin marketplace add RobsonTigre/everyday-causal-skills`, then `/plugin install everyday-causal-skills@everyday-causal-skills` and `/reload-plugins`. To verify, `/causal-planner` should ask about your causal question.
Does the set work in Codex and Cursor?
Yes. For Codex copy the folders from `skills/` into `~/.agents/skills/` and restart it. For Cursor clone the repository into `~/.cursor/plugins/local/`. There is also a Gemini CLI extension and a Copilot CLI plugin.
Which method fits a launch in some of the stores?
With treated and untreated groups and data before and after, the planner recommends difference-in-differences and hands off to `/causal-did`. The skill first checks whether the groups followed similar trends before launch.
What does `/causal-auditor` check?
It checks a finished analysis against five categories of validity threats: whether something else could explain the effect and whether group selection is biased. The output is a list of threats to address before presenting.
What is `/causal-exercises` for?
For practice: the skill gives simulated data with a known true effect and gives feedback on your chosen approach.
Discussion
No questions or reviews yet.
Related
PluginsAuto-Empirical Research Skills
Auto-Empirical Research Skills (AERS)
A plugin marketplace for empirical social science research: causal inference, econometrics and manuscript prep for top journals
17 skills for economists: research ideas, literature review, Stata and R, LaTeX models, tables and Beamer slides
High-Stakes Analytics Lab
High-Stakes Analytics & Decision Lab
An evidence-gated analytics skill for high-stakes decisions: profiles data, picks a method the evidence supports, can answer with a data request
A marketplace of 16 Claude Code plugins for quantitative and qualitative sociology research, with pauses between phases.
31 skills for Claude Code and Codex for the analyst workflow: data checks, cohorts, funnels, A/B tests, dashboards, summaries
Probabl skills for machine learning experiments: a scikit-learn and skrub pipeline, evaluation, audit and a link to Skore Hub
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Everyday Causal Skills (this card) | no | 84 | ||
| StatsPAI: causal inference and econometrics for agents | no | not required | 374 | the MCP server and skill of the StatsPAI library: the agent runs econometric methods directly |
| Auto-Empirical Research Skills | no | 5,007 | a marketplace for empirical social science research with a broader set of tasks |
README in short
The description starts with whether a change helped or circumstances coincided. Two worked examples show A/B test design and loyalty program evaluation. A skills table, a five-step flow, warnings about how `/causal-dag` and `/causal-roi` differ and install steps for five environments follow.