research-discipline SKILL.md skillA

research-discipline from karsten-s-nielsen/mad-scientist-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add karsten-s-nielsen/mad-scientist-skills/research-discipline@git:20261001.d0fd4dd

Pins git:20261001.d0fd4dd in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
karsten-s-nielsen/mad-scientist-skills
Path
plugins/mad-scientist-skills/skills/research-discipline/SKILL.md
Versions recorded
1
Latest version
git:20261001.d0fd4dd, committed Oct 1, 2026
Size
6,969 bytes
Repository stars
3
License
MIT

Use when designing or validating a quantitative metric, model, or ranking, when pre-registering an analysis, or before shipping a research claim or write-up to a paper, deck, post, or proposal — pre-registers the E1–E7 metric-design anti-patterns, the validity ladder (construct → face → predictive → robustness), and the multi-pass review protocol. Also use when scrutinising a research metric or empirical claim for soundness. Triggers on "design a metric", "validate this metric/model/ranking…

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

# Research discipline

A portable discipline for building, validating, reviewing, and shipping quantitative football-analytics
claims — metrics, models, rankings, research write-ups. Adopt it as a **pre-registration gate**: before
fitting or comparing anything, walk the anti-patterns and the validity ladder and record your answers; use
the multi-pass protocol for reviews and the defect smell-test as a pre-ship check.

It is a **checklist, not a rulebook** — every threshold and exclusion is a *documented, defensible choice*,
not a law. When you deviate, say so and why, and keep the completed checklist with the work.

The E1–E6 taxonomy below is from Rahimian (2026); E7 is our own observability extension. Full citations and
provenance are in [`references/sources.md`](references/sources.md).

## Metric-design anti-patterns (E1–E7)

A metric proposal must clear all seven before it is called validated. Each has a published or in-house anchor.

- **E1 — Learning the provider's label instead of measuring the construct.** If the supervised target *is* a
  vendor's event label (or it leaks in as a feature), you have re-implemented the vendor, not measured the
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add karsten-s-nielsen/mad-scientist-skills/research-discipline@sha256:4461a2f93dd11b6b

The line at the top of this page pins the label git:20261001.d0fd4dd, 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 badge

[![mdr](https://markdownregistry.com/badge/art_ebsxm3n3nircs3tx.svg)](https://markdownregistry.com/a/art_ebsxm3n3nircs3tx)

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

Versions

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

versioncommittedcommitsizeaudit
git:20261001.d0fd4dd latest2026-10-01 d0fd4dd 6,969 BA 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 (6969 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

karsten-s-nielsen/mad-scientist-skills · 3 stars · license MIT · pushed 2026-10-01 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_ebsxm3n3nircs3tx
GET https://markdownregistry.com/api/v1/resolve?ref=karsten-s-nielsen/mad-scientist-skills/research-discipline
GET https://markdownregistry.com/api/v1/blob/4461a2f93dd11b6b0d1c1adf4f41d95e87d4e7b98be8bbc266a1e209c60dafcc

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.

More from karsten-s-nielsen/mad-scientist-skills

AGENTS.md AGENTS.md file
karsten-s-nielsen/mad-scientist-skills · AGENTS.md
git:20260926.35dafdd · audit A · 3 stars
CLAUDE.md CLAUDE.md file
karsten-s-nielsen/mad-scientist-skills · CLAUDE.md
git:20260926.35dafdd · audit B · 3 stars
architecture-audit SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/architecture-audit/SKILL.md · Comprehensive architecture audit with two modes and a single tier (beta). Planning mode evaluates architectural…
git:20260926.35dafdd · audit A · 3 stars
c4 SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/c4/SKILL.md · Creates interactive C4 model architecture diagrams using Structurizr DSL — the official C4 notation by Simon Brown…
git:20260813.1e1da60 · audit B · 3 stars
cognitive-interface-audit SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/cognitive-interface-audit/SKILL.md · Comprehensive cognitive interface audit with two modes and a single tier. Planning mode designs task models, user…
git:20260926.35dafdd · audit A · 3 stars
documentation-audit SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/documentation-audit/SKILL.md · Comprehensive documentation audit with two modes and a single tier. Planning mode designs documentation strategy…
git:20260926.35dafdd · audit A · 3 stars
final-review SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/final-review/SKILL.md · Pre-commit quality gate — reviews all code and documentation for consistency, best practices, and completeness, then…
git:20260929.7b74d3a · audit A · 3 stars
measure-before-optimize SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/measure-before-optimize/SKILL.md · Pre-change measurement gate for perf-sensitive functions. Use BEFORE modifying any function that has a pytest-benchmark…
git:20260929.7b74d3a · audit A · 3 stars
observability-audit SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/observability-audit/SKILL.md · Comprehensive observability audit with two modes and two tiers (beta). Planning mode designs telemetry strategy…
git:20260416.d18ed8c · audit A · 3 stars
optimization-audit SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/optimization-audit/SKILL.md · Comprehensive optimization audit with two modes and a single tier. Planning mode designs performance strategy, capacity…
git:20260929.7b74d3a · audit B · 3 stars
security-audit SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/security-audit/SKILL.md · Comprehensive security audit with two modes and two tiers. Planning mode reviews architecture for security gaps (STRIDE…
git:20260724.35d8a81 · audit B · 3 stars
unbiased-review SKILL.md skill
karsten-s-nielsen/mad-scientist-skills · plugins/mad-scientist-skills/skills/unbiased-review/SKILL.md · Use when reviewing a spec, plan, or implementation produced by another session - verifies its claims against the actual…
git:20260929.7b74d3a · audit A · 3 stars

Every file in karsten-s-nielsen/mad-scientist-skills

Other files named research-discipline

research-discipline SKILL.md skill
hkuds/vibe-trading · agent/src/skills/research-discipline/SKILL.md · A short self-bias checklist to run at the START of any investment research task (stock screen / sector study / company…
git:20260706.6c09d25 · audit A · 36,037 stars

Browse by kind, by grade A, or by owner.