Home / cuhk-aim-group / neurodiscovery · GitHub

causal-treatment-models SKILL.md skillA

causal-treatment-models from cuhk-aim-group/neurodiscovery is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned

mdr add cuhk-aim-group/neurodiscovery/causal-treatment-models@git:20260825.cbd02ff

Pins git:20260825.cbd02ff in mdr.lock. Pin the fingerprint instead.

Audit grade: A, 17 of 17 checks passed
Kind: SKILL.md skill
Repository: cuhk-aim-group/neurodiscovery
Path: skills/causal-treatment-models/SKILL.md
Versions recorded: 1
Latest version: git:20260825.cbd02ff, committed Aug 25, 2026
Size: 4,362 bytes
Repository stars: 93
License: MIT

Use this skill whenever the scientific target is a treatment effect rather than ordinary outcome prediction: propensity weighting, S/T/X learners, doubly robust learning, policy learning, causal forests, TARNet, DragonNet, CATE estimation, heterogeneous treatment effects, and individualized treatment selection. Triggers include 'causal inference', 'treatment effect', 'CATE', 'ATE', 'propensity score', 'IPW', 'doubly robust', 'causal forest', 'TARNet', 'DragonNet', and 'treatment policy'.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Causal Treatment Models Skill

## Overview

`causal-treatment-models` estimates average or conditional treatment effects
from observational subject-level features. It is for the contrast
`Y(1) - Y(0)`, not for predicting the observed outcome alone.

**Supported estimators**

| Model | Output |
|---|---|
| `ipw` | propensity-weighted ATE as constant CATE |
| `s_learner` | single outcome model treatment contrast |
| `t_learner` | separate treated/control outcome models |
| `x_learner` | imputed effects blended by propensity |
| `doubly_robust` | doubly robust pseudo-outcome CATE |
| `policy_learner` | interpretable treatment assignment policy |
| `causal_forest` | `econml` CausalForestDML |
| `tarnet` | shared representation with two outcome heads |
| `dragonnet` | TARNet plus propensity head |

The CLI uses cross-fitted held-out predictions. Causal interpretation still
requires consistency, positivity, no unmeasured confounding, and a defensible
temporal ordering.

---

## Installation

```bash
pip install numpy pandas scipy scikit-learn joblib torch
```

For Causal Forest:

```bash
pip install econml
```

---

## Workflows

### 1. Prepare treatment data

```text
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint

mdr add cuhk-aim-group/neurodiscovery/causal-treatment-models@sha256:b2292cd81f3a35d5

The line at the top of this page pins the label git:20260825.cbd02ff, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

[![mdr](https://markdownregistry.com/badge/art_pzb4iy35v4yc7fmz.svg)](https://markdownregistry.com/a/art_pzb4iy35v4yc7fmz)

5 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on Aug 25, 2026.

version	committed	commit	size	audit
git:20260825.cbd02ff latest	2026-08-25	cbd02ff	4,362 B	A	view

Audit of the latest version

A 17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.

pass: Frontmatter block present
pass: Frontmatter declares a name
pass: Frontmatter declares a description
pass: Size between 200 bytes and 200 KB (4362 bytes)
pass: No zero-width or bidi control characters
pass: No instruction hidden inside an HTML comment
pass: No link to an exfiltration or paste host
pass: No credential-shaped string
pass: No instruction to send local credentials anywhere
pass: No text hidden with inline styles
pass: No prompt-injection phrasing
pass: No curl or wget piped into a shell
pass: No recursive delete of root, home or parent
pass: No instruction to read or print local credentials
pass: No base64 blob over 200 characters
pass: No link to a raw IP address
pass: No script tag

Source

GitHub

cuhk-aim-group/neurodiscovery · 93 stars · license MIT · pushed 2026-10-23 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_pzb4iy35v4yc7fmz
GET https://markdownregistry.com/api/v1/resolve?ref=cuhk-aim-group/neurodiscovery/causal-treatment-models
GET https://markdownregistry.com/api/v1/blob/b2292cd81f3a35d55ee9e552cd6c0d7a1554077572257523865831aa3e4bcd2b

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.