workflow-redesign SKILL.md skillA

workflow-redesign from lugassawan/swe-workbench is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add lugassawan/swe-workbench/workflow-redesign@git:20261006.bf03fb1

Pins git:20261006.bf03fb1 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
lugassawan/swe-workbench
Path
skills/workflow-redesign/SKILL.md
Versions recorded
1
Latest version
git:20261006.bf03fb1, committed Oct 6, 2026
Size
12,775 bytes
Repository stars
3
License
MIT

Replace the approach of a PR — weigh redesign alternatives against its current design, then reimplement or supersede after approval. Activated by /swe-workbench:design --pr.

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

Announce at start: "Using `swe-workbench:workflow-redesign` — re-planning the approach of an open PR."

## When to invoke

- `/swe-workbench:design --pr [N] "<why>"` — the command parses flags and dispatches here
- The PR works but uses the wrong abstraction, too many moving parts, or the wrong layer

## When NOT to invoke

- Adding a related sub-idea to the PR → `/swe-workbench:extend`
- Keeping the design and cleaning it up → `/swe-workbench:refactor`
- Fixing review comments → `/swe-workbench:address-feedback`
- A PR you do not own → `/swe-workbench:review`

## Precondition

The command passes `PR_ARG` (optional PR number), `WHY` (the user's reason), `NEW_PR` (`true` when `--new-pr` was given, else `false`) and `MODE` (`grill`, `standard`, or empty when the invocation carried no explicit signal). If `WHY` is empty, ask for it once before continuing.

Every abort or stop path in Phases A–F still runs the `$RUN_DIR` reap (`swe-workbench-reap-run-dir "$RUN_DIR"`, once it exists) and, if `$WT` exists, prints its path. Nothing is ever removed on an abort.

## Phase A — Read-only context

Nothing in this phase writes to the repo or the PR.

```bash
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add lugassawan/swe-workbench/workflow-redesign@sha256:5be227ce7c20c4ec

The line at the top of this page pins the label git:20261006.bf03fb1, 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_rv2mvrclqfg7pllf.svg)](https://markdownregistry.com/a/art_rv2mvrclqfg7pllf)

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

Versions

The registry has recorded 1 version of this file, committed on Oct 6, 2026.

versioncommittedcommitsizeaudit
git:20261006.bf03fb1 latest2026-10-06 bf03fb1 12,775 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 (12775 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

lugassawan/swe-workbench · 3 stars · license MIT · pushed 2026-11-07 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_rv2mvrclqfg7pllf
GET https://markdownregistry.com/api/v1/resolve?ref=lugassawan/swe-workbench/workflow-redesign
GET https://markdownregistry.com/api/v1/blob/5be227ce7c20c4ec2a6f6e21a217bdf0fe4d5dea0213814c9e15455f6ec6a167

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 lugassawan/swe-workbench

comms-context SKILL.md skill
lugassawan/swe-workbench · skills/comms-context/SKILL.md · Fetch on-call and discussion context from Slack threads and PagerDuty incidents — participants, timeline, and status…
git:20260828.71a6399 · audit A · 3 stars
language-bash SKILL.md skill
lugassawan/swe-workbench · skills/language-bash/SKILL.md · Bash idioms — strict mode, parameter expansion, arrays, trap cleanup, heredocs, and POSIX portability. Auto-load when…
git:20260828.71a6399 · audit A · 3 stars
language-csharp SKILL.md skill
lugassawan/swe-workbench · skills/language-csharp/SKILL.md · C# and .NET idioms — .NET 8 LTS, C# NRT, records, value semantics, switch expressions, async/await, Task, ValueTask…
git:20260828.71a6399 · audit A · 3 stars
language-dart SKILL.md skill
lugassawan/swe-workbench · skills/language-dart/SKILL.md · Dart idioms — null safety, Futures and Streams, widget composition, and Riverpod app architecture. Auto-load when…
git:20260828.71a6399 · audit A · 3 stars
language-go SKILL.md skill
lugassawan/swe-workbench · skills/language-go/SKILL.md · Go idioms — error handling, concurrency, and standard library usage. Auto-load when working with .go files, go.mod…
git:20260828.71a6399 · audit A · 3 stars
language-java SKILL.md skill
lugassawan/swe-workbench · skills/language-java/SKILL.md · Java idioms — records, sealed types, virtual threads, and JDK 21+ patterns. Auto-load when working with .java files…
git:20260828.71a6399 · audit A · 3 stars
language-kotlin SKILL.md skill
lugassawan/swe-workbench · skills/language-kotlin/SKILL.md · Kotlin idioms — null safety, sealed interfaces, scope functions, and Flow. Auto-load when working with .kt files…
git:20260828.71a6399 · audit A · 3 stars
language-python SKILL.md skill
lugassawan/swe-workbench · skills/language-python/SKILL.md · Python idioms — PEP 8, context managers, generators, asyncio, and testing. Auto-load when working with .py files…
git:20260828.71a6399 · audit A · 3 stars
language-ruby SKILL.md skill
lugassawan/swe-workbench · skills/language-ruby/SKILL.md · Ruby idioms — Ruby 3.x pattern matching, procs, Comparable, error handling, and testing. Auto-load when working with…
git:20260828.71a6399 · audit A · 3 stars
language-rust SKILL.md skill
lugassawan/swe-workbench · skills/language-rust/SKILL.md · Rust idioms — ownership, lifetimes, and iterators. Auto-load when working with .rs files, Cargo.toml, or when the user…
git:20260828.71a6399 · audit A · 3 stars
language-sql SKILL.md skill
lugassawan/swe-workbench · skills/language-sql/SKILL.md · SQL idioms — query tuning, table definitions, constraints, transactions, and pagination. Auto-load when working with…
git:20260828.71a6399 · audit A · 3 stars
language-swift SKILL.md skill
lugassawan/swe-workbench · skills/language-swift/SKILL.md · Swift idioms — optionals, value types, async/await, and Result builders. Auto-load when working with .swift files…
git:20260828.71a6399 · audit A · 3 stars

Every file in lugassawan/swe-workbench

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