agentleFS
HermeticOrmusHermeticOrmus/dx-audit-skills/skills/dx-audit/SKILL.md

dx-audit

Skill0 starsChanged 6 months agoMITView on GitHub

Audit and improve developer experience in a codebase by inspecting README, quickstart, scripts, contributing guide, env docs, error messages, and tooling, scoring onboarding friction, and reporting findings or implementing fixes. Use when a new contributor hits friction, when setup is undocumented, or when the README has not been run on a clean checkout since the stack changed.

  • Reads credentials

6 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions6 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 6

    1. dotslashsimran/simcraft★ 451same name
    2. mblode/agent-skills★ 203same name
    3. jmcelreavey/devhub★ 5same name
    4. nnennandukwe/skills★ 4same name
    5. SgtPooki/skills★ 1same name
    6. SWEStash/swe-workflow-skills★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Default to report-only when the mode is unstated, because applying changes a contributor never reviewed is itself a DX failure.

    line 51

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

DX audit

A protocol for auditing and improving developer experience in a codebase. Developer experience is everything a contributor touches before their first useful line of code. High onboarding friction is silent: no one files a bug for it, every new contributor just pays the cost.

Usage

/dx-audit                 # full audit of the current project (report-only)
/dx-audit [path]          # audit a specific directory
/dx-audit --report-only   # inspect and report, change nothing (default)
/dx-audit --implement     # apply fixes with approval, verify each

What to inspect

Walk the codebase as a new contributor would. For each dimension, record what exists, what is missing, and whether what exists is accurate.

  1. README: first-paragraph clarity, a copy-pasteable quickstart, current commands, links to deeper docs.
  2. Quickstart and setup: the path from clone to running app, in dependency order, with prerequisites listed. Run it on a clean checkout and count what fails.
  3. Scripts: install / dev / build / test / lint / format entries in a task runner, with names that match the README and contributing guide.
  4. Contributing guide: branch naming, commit format, PR expectations, how to run tests, checked against the actual workflow.
  5. Environment docs: an .env.example listing every required variable, each annotated with what it is and where to get it.
  6. Error messages: setup scripts fail loudly with actionable messages, so missing deps, wrong runtime version, or absent env vars produce a sentence, not a stack trace.
  7. Tooling: linter, formatter, type-checker configured and wired to a pre-commit hook or CI so they run without being remembered.

How to score friction

Assign each dimension a level. The goal is to find where the silent cost concentrates.

  • None: present, accurate, discoverable. A new contributor succeeds without asking.
  • Low: present but incomplete or stale. Costs a minute of guessing.
  • Medium: present but wrong, or absent where it matters. Costs a question to a maintainer.
  • High: absent or actively misleading. Blocks a new contributor until someone intervenes.

Anchor the score with two timings: time to first run (clone to running app or passing test) and time to first contribution (clone to a tested, formatted, ready-to-open PR).

Report-only vs implement modes

State the active mode before producing output.

Report-only (default): inspect, score, and produce a prioritized findings list grouped by effort (quick wins, medium effort, larger work). Each finding names the dimension, the friction level, and the specific fix. Change nothing.

Implement: apply fixes with approval, highest-friction and lowest-effort first. Make surgical changes: add the missing section, create the .env.example, add the script, fix the message. Verify each fix on a clean checkout before claiming it done. Stop after each group and report what changed and what was verified.

Default to report-only when the mode is unstated, because applying changes a contributor never reviewed is itself a DX failure.

Verify before claiming a fix

Run the quickstart on a clean checkout, not on your already-configured machine. Run each documented script. Trigger the missing-env-var and wrong-version paths and read the message a contributor would see. A README that has not been run on a clean checkout is a guess, not a fix.


See full content at https://github.com/HermeticOrmus/dx-audit-skills.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.