workflow-audit plugin
A Claude Code plugin for SwiftUI that walks an app like a user and finds unreachable screens, dead ends and placeholder data.
- 60 ★ on the repository
- Last commit
- Apache-2.0
Extended access
Why this level
- The plugin reads project source and writes reports without running anything.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/workflow-auditDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The plugin reads project source and writes reports without running anything.
This is third-party code. Review the repository files before installing.
What it does
workflow-audit checks paths between files in a SwiftUI app rather than single lines. An audit starts at every door into a feature, such as a button, menu item or card, and follows it forward. It looks for screens nobody can reach and numbers that code computes but no screen shows. Checks run in five layers: `layer1` finds all entries and features with no way in, the second layer traces the path, the third catches 33 kinds of problems (dead ends, screens that will not close, buttons below the fold, spinners with no timeout, leftover fake data), the fourth judges how it feels to a newcomer and the fifth finds placeholder data. A second skill, `/plan --workflow-audit`, turns findings into an ordered plan with no ready code.
Who it is for. SwiftUI developers who want to confirm before release that every feature they built can be found and finished by a user.
Good fit when
- A finished screen has no button that opens it and the feature effectively does not exist.
- The app computes a number but the screen shows a blank space or a stale value.
- Before release you need to walk the app through a new user eyes: how many taps and whether anything confirms.
- After the audit you need a fix order with the cost of each phase and a way to undo it.
Not a fit when
- You need a check on every file save: the audit costs more than a linter and is meant for release time.
- You need to know whether a handler does the right thing: the plugin reads code and does not run it.
- The project is not SwiftUI: the check targets that stack.
Example request
/workflow-audit layer1 and show every way into my SwiftUI app features together with features that cannot be reached.Limitations
The plugin reads code and does not run the app, so it cannot see wrong but reachable behaviour or correctly wired but wrongly drawn controls. A full five layer audit takes real time and spends tokens, with `layer1` the cheapest at about ten minutes. It complements a linter and does not replace one: style and pattern problems are better caught on every save. Enter the commands in Claude Code one at a time, otherwise the second line reads as part of the first.
How to disable. In Claude Code run `/plugin uninstall workflow-audit@workflow-audit` and remove the added marketplace if needed. Report files in the project can be kept or deleted.
FAQ
Which layer should I start with?
With `/workflow-audit layer1`: about ten minutes and you get a list of every way into features. Check it against your own idea of the app.
What problems does the third layer catch?
33 kinds: dead ends, screens that will not close, buttons below the fold, spinners with no timeout and fake data left in.
How is this different from a linter?
A linter checks each file separately on every save. The audit checks connections between files and runs before a release.
How do I get a fix plan?
After the audit run `/plan --workflow-audit`. The plan is ordered, names the file and line range and a done condition, and gives no code to paste.
Does the plugin run the app?
No, it reads source code. So it cannot judge whether a handler does the right thing or whether the screen looks as intended.
Discussion
No questions or reviews yet.
Related
PluginsA Claude Code plugin for UX regression testing that walks user journeys in a live app and returns a ranked fix plan.
A six-skill accessibility plugin: page review, WCAG 2.2 audit with Playwright, audit planning and an improvement roadmap.
Rails Audit by thoughtbot
Rails Audit (thoughtbot)
A thoughtbot skill for reviewing Rails applications, finding code and test issues and ranking findings by severity
show-your-work
claude-show-your-work
A plugin that shows images and video in the Claude Code terminal, records Playwright checks and compares before and after
A Claude Code plugin that verifies a change against its plan: boots your stack, runs acceptance criteria, reports evidence.
A Claude Code plugin set: Figma visual regression, test and code review, project lore and session analysis as an HTML report
README in short
The project includes five audit layers and a separate fix planner. It solves one problem: showing where a user fails to reach a result even though every code line is correct.