Home / pedrohcgs / claude-code-my-workflow · .claude/skills/differential-audit/SKILL.md · GitHub

differential-audit skillA

differential-audit is agent-read markdown (skill) from pedrohcgs/claude-code-my-workflow: Compare two implementations of the same thing — a port (R↔Python↔Stata), a reimplementation, a replication package, a refactor, or a new version against the old — so that agreement means something. Freeze inputs first, inventory every expected output, test the comparator itself, compare every channel (not just the headline number), and give each divergence a stable ID and a smallest witness. Use for cross-language parity, replication, upgrade/regression gates, or whenever "the numbers match" is .

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified before it reaches your agent: the main file against the SHA-256 recorded here, the others against the git hashes of its source commit. The deterministic audit below grades the latest version, and the same checks always give the same file the same grade.

What the file says

# Make agreement mean something

Two implementations agreeing proves they satisfy a **prespecified contract**. It does not prove either is correct, and it never validates the method or its assumptions. Both can be wrong in the same way — especially when one was written by reading the other. Design the comparison so that agreement is informative and disagreement is legible.

**Rule: freeze before you compare; test the comparator before you trust it.**

## 1. State the claim and the reference

Write down: what is being compared, which side is the reference, and **what agreement would and would not establish**. "Matches the R package" is a conformance claim, not a correctness claim. Say so explicitly, so nobody later reads parity as validation.

## 2. Freeze the inputs before inspecting anything

Record and fix: data versions or hashes, code and package versions, random seeds or realized sample splits, options and defaults, the outputs to be compared, and the acceptance thresholds. Freezing *after* a first look invites tolerance drift toward whatever the run produced.
…

Read the whole file at its exact version.

How to install

Latest version
mdr add pedrohcgs/claude-code-my-workflow/differential-audit@git:20260822.d67758f
Exact content
mdr add pedrohcgs/claude-code-my-workflow/differential-audit@sha256:72335e59883f8fcf

Pin to a label to follow the author's releases, or to a sha256 for exact bytes. Either way the resolved hash is written to mdr.lock, and mdr install fetches those bytes again and checks them, so it installs them exactly or fails.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_ji2c6vfdcm7vxiob.svg)](https://markdownregistry.com/a/art_ji2c6vfdcm7vxiob)

2 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260822.d67758f latest2026-08-22 d67758f 6,248 BA view · diff
git:20260821.66003332026-08-21 6600333 6,253 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, 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 (6248 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

pedrohcgs/claude-code-my-workflow · 1,598 stars · license MIT · pushed 2026-09-27 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_ji2c6vfdcm7vxiob
GET https://markdownregistry.com/api/v1/resolve?ref=pedrohcgs/claude-code-my-workflow/differential-audit
GET https://markdownregistry.com/api/v1/blob/72335e59883f8fcf285a1cd4af475c7ecda4aa194302a60475b3715a26a59167

Your agent 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.

More from pedrohcgs/claude-code-my-workflow

adjudicate-review skill
pedrohcgs/claude-code-my-workflow · .claude/skills/adjudicate-review/SKILL.md · Turn an incoming set of findings — from an AI reviewer, a referee report, a code review, a linter, or a second model —…
git:20260927.6535de7 · audit A · 1,598 stars
audit-reproducibility skill
pedrohcgs/claude-code-my-workflow · .claude/skills/audit-reproducibility/SKILL.md · Enforce the replication-protocol.md rule by cross-checking numeric claims in a manuscript against the actual R / Stata…
git:20260927.6535de7 · audit A · 1,598 stars
blast-radius skill
pedrohcgs/claude-code-my-workflow · .claude/skills/blast-radius/SKILL.md · Before and after changing anything shared — a function's return value, a signature, a schema, a label set, a config…
git:20260823.6b8fa14 · audit A · 1,598 stars
capture-environment skill
pedrohcgs/claude-code-my-workflow · .claude/skills/capture-environment/SKILL.md · Snapshot the computational environment for a replication package — detects the analysis stack (R / Stata / Python) and…
git:20260927.6535de7 · audit A · 1,598 stars
challenge skill
pedrohcgs/claude-code-my-workflow · .claude/skills/challenge/SKILL.md · Stress-test a finding against the choices you did not make. Enumerates the discrete forks a competent analyst could…
git:20260823.6b8fa14 · audit A · 1,598 stars
checkpoint skill
pedrohcgs/claude-code-my-workflow · .claude/skills/checkpoint/SKILL.md · Save a structured state snapshot before stopping or handing off. Captures the active plan, recent decisions, file…
v1.0.0 · audit A · 1,598 stars
coauthor-brief skill
pedrohcgs/claude-code-my-workflow · .claude/skills/coauthor-brief/SKILL.md · Generate a co-author / collaborator handoff brief for a multi-author, multi-machine project — summarizing what changed…
v1.0.0 · audit A · 1,598 stars
commit skill
pedrohcgs/claude-code-my-workflow · .claude/skills/commit/SKILL.md · Stage, commit, push, open a PR, and merge to main. Use ONLY on explicit commit intent — user says "commit", "ship it"…
git:20260927.6535de7 · audit A · 1,598 stars
compile-latex skill
pedrohcgs/claude-code-my-workflow · .claude/skills/compile-latex/SKILL.md · Compile a Beamer LaTeX slide deck with XeLaTeX (3 passes + bibtex). Use when user says "compile", "build the slides"…
git:20260415.3ef908f · audit A · 1,598 stars
compress-session skill
pedrohcgs/claude-code-my-workflow · .claude/skills/compress-session/SKILL.md · Distill the current conversation into a structured note (decisions made, open questions, file pointers with line…
v1.0.0 · audit A · 1,598 stars
context-status skill
pedrohcgs/claude-code-my-workflow · .claude/skills/context-status/SKILL.md · Show current context status and session health. Use to check how much context has been used, whether auto-compact…
v1.0.0 · audit A · 1,598 stars
create-lecture skill
pedrohcgs/claude-code-my-workflow · .claude/skills/create-lecture/SKILL.md · Create a new Beamer lecture `.tex` from source papers and materials, with notation consistency checks and the project's…
git:20260927.6535de7 · audit A · 1,598 stars

Every file in pedrohcgs/claude-code-my-workflow

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