Marginaleffects

What it does

When your agent uses it

• Users ask about predictions
• Marginal effects
• Average treatment effects (ATE/ATT/CATE)
• Hypothesis testing

Example prompts

• “Model to Meaning”
• “/marginaleffects”

Requirements

• Python 3
• Pre-approved tools (allowed-tools): Read, Grep, Glob

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

1. 1Quantity: What estimand? (predictions, comparisons, slopes, or tests)
2. 2Predictors (Grid): Where to evaluate? (observed values, counterfactual scenarios, balanced grids)
3. 3Aggregation: Over whom? (unit-level, group means with by=, weighted averages)
4. 4Uncertainty: Which inference method? (delta method, robust SE, bootstrap, Bayesian)
5. 5Test: What hypothesis? (null tests, equivalence, pairwise contrasts)

What it can do on your machine

Read from SKILL.md and the folder at commit 9fa87d8. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Grep
  • Glob

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are r and python).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • marginaleffects.com
  • routledge.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Marginaleffects loads about 2.1k tokens when it runs. Until then it costs about 132 tokens; SKILL.md has 702 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~132

When it runs · the whole SKILL.md, loaded when a task matches

~2.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from brycewang-stanford/Auto-Empirical-Research-Skills at commit 9fa87d8, republished under its CC-BY-4.0 licence (© brycewang-stanford). 702 words, ~2,108 tokens.

name

marginaleffects

description

Manual for the marginaleffects R and Python package, and guide to the book "Model to Meaning". Use when users ask about predictions, comparisons, slopes, marginal effects, average treatment effects (ATE/ATT/CATE), hypothesis testing, contrasts, counterfactuals, risk ratios, odds ratios, causal inference with G-computation, or need help with marginaleffects functions like predictions(), comparisons(), slopes(), hypotheses(), datagrid(), avg_predictions(), avg_comparisons(), avg_slopes(), or plot functions.

allowed-tools

Read, Grep, Glob

license

CC-BY-4.0

metadata.source

https://marginaleffects.com

metadata.maintainer

vincentarelbundock

library(marginaleffects)

# Fit logistic regression
mod <- glm(am ~ hp + wt, data = mtcars, family = binomial)

# Average marginal effects (slopes on probability scale)
avg_slopes(mod)

# Predicted probabilities at specific values
predictions(mod, newdata = datagrid(hp = c(100, 150, 200), wt = 3))

# Average treatment effect: compare hp = 150 vs hp = 100
avg_comparisons(mod, variables = list(hp = c(100, 150)))

# Risk ratio for a 50-unit increase in hp
avg_comparisons(mod, variables = list(hp = 50), comparison = "ratio")

import marginaleffects as me
import statsmodels.formula.api as smf

# Fit logistic regression
mod = smf.logit("am ~ hp + wt", data=me.get_dataset("mtcars")).fit()

# Average marginal effects
me.avg_slopes(mod)

# Predicted probabilities at specific values
me.predictions(mod, newdata=me.datagrid(mod, hp=[100, 150, 200], wt=3))

# Average treatment effect: compare hp = 150 vs hp = 100
me.avg_comparisons(mod, variables={"hp": [100, 150]})

© brycewang-stanford, CC-BY-4.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/39-vincentarelbundock-marginaleffects of brycewang-stanford/Auto-Empirical-Research-Skills.

• SKILL.md
• README-original.md

Open the folder on GitHubat commit 9fa87d8