agentleFS
Sign inStart free
ldaumeldaume/agentic-engineering-harness/skills/design/design-review/SKILL.md

design-review

Skill3 starsChanged 3 months agoMITView on GitHub

Use whenever auditing, critiquing, or verifying an existing interface, website, user flow, form, dashboard, slide deck, or diagram - UX audit, accessibility audit, heuristic review, design critique, visual QA, browser verification with Playwright and axe, and turning findings into a prioritized improvement backlog (UX-Audit, Design-Review, Barrierefreiheit prüfen, Usability-Review). Every finding names an observation, impact, evidence, and how to validate it. Not for reviewing code correctness, which code review owns. Fixing what it finds belongs to frontend-craft.

155 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use project-stage-detect, team-audio and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • CoralGameSkill

    project-stage-detect

    Automatically analyze project state, detect stage, identify gaps, and recommend next steps based on existing artifacts.

    Also in 16 projects that use this

  • CoralGameSkill

    team-audio

    Orchestrate audio team: audio-director + sound-designer + technical-artist + gameplay-programmer for full audio pipeline from direction to implementation.

    Also in 13 projects that use this

  • julianoczkowskiSkill

    brief-to-tasks

    Break a design brief into an ordered checklist of independently buildable tasks using vertical slices. Saves as a markdown checklist. Use when user wants to break down work, create tasks from a brief, plan implementation order, or mentions "tasks" or "breakdown".

    Also in 7 projects that use this

  • lilangMaxSkill

    asset-spec

    Generate per-asset visual specifications and AI generation prompts from GDDs, level docs, or character profiles. Produces structured spec files and updates the master asset manifest. Run after art bible and GDD/level design are approved, before production begins.

    Also in 10 projects that use this

  • lilangMaxSkill

    create-control-manifest

    After architecture is complete, produces a flat actionable rules sheet for programmers — what you must do, what you must never do, per system and per layer. Extracted from all Accepted ADRs, technical preferences, and engine reference docs. More immediately actionable than ADRs (which explain why).

    Also in 10 projects that use this

  • +4 more from CoralGame
  • +3 more from lilangMax

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 129 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run the automated gate (references/browser-verification.md): axe WCAG 2.2 A/AA, overflow at key widths, console errors, screenshots per …

    line 20
  2. Distinctive:

    Check the artifact-specific list from its owner skill (forms, flows, product UI, web, dashboards, slides and diagrams, design system).

    line 24
  3. Very specific:

    Write each finding in the format of references/review-method.md, with a classification and confidence.

    line 26
  4. Distinctive:

    Every finding has location, observation, impact, evidence, confidence, severity, recommendation, and validation.

    line 54

“Common” is measured: kept by at least 6 in 10 of the 129 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

Design Review

Find what stops people from reaching their goal, show the evidence, and turn it into changes ranked by impact. "The design feels dated" helps nobody; "the primary action sits below the fold at 360 px, so mobile visitors miss it" can be fixed and checked.

Golden path

  1. Name the person, the task, and the journey under review; a review without a task becomes a taste contest.
  2. Walk the task end to end in the real product, desktop and mobile, with keyboard only, and with a screen reader for critical paths.
  3. Run the automated gate (references/browser-verification.md): axe WCAG 2.2 A/AA, overflow at key widths, console errors, screenshots per state. Automation finds candidates, not conformance: axe states it finds about 57% of WCAG issues (verified) [SRC-TOOL-001].
  4. Check the artifact-specific list from its owner skill (forms, flows, product UI, web, dashboards, slides and diagrams, design system).
  5. Write each finding in the format of references/review-method.md, with a classification and confidence.
  6. Prioritize by user impact, frequency, severity, journey criticality, accessibility risk, effort, uncertainty, and learning value.
  7. Hand over a backlog: smallest valuable change first, each with its validation.

Rules that change the result

  • Separate what was observed from what is inferred; a likely problem is labeled as such until a test or data confirms it.
  • An accessibility violation cites the WCAG 2.2 success criterion (verified) [SRC-A11Y-001]; WCAG 3 is a working draft and is not a conformance basis (verified) [SRC-A11Y-002].
  • Aesthetic findings state the observation, the effect on the person, and an alternative; otherwise they are dropped.
  • A green automated run never closes the review; the manual list in references/browser-verification.md stays open until someone checks it.

Load when

  • references/review-method.md - always: dimensions, finding format, classification, severity, prioritization, backlog.
  • references/browser-verification.md - whenever the product can be opened in a browser.

Done when

Every finding has location, observation, impact, evidence, confidence, severity, recommendation, and validation; the automated gate ran or its absence is stated; the manual checks are done or assigned.

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.

Check your setup, free