agentleFS
xingfanxiaxingfanxia/AX-skills/plan-design-review/SKILL.md

plan-design-review

Skill4 starsChanged 55 days agoMITView on GitHub

Review the design completeness of an implementation plan when the user asks for a plan design review or design-completeness check. Assess seven dimensions, concrete interaction states, and generic design risks. Review-only requests produce findings; requested plan improvements apply clear fixes and batch consequential choices.

49 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use plan-tune, ios-clean 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.
  • ShreyRaviSkill

    plan-tune

    Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). Review which AskUserQuestion prompts fire across gstack skills, set per-question preferences (never-ask / always-ask / ask-only-for-one-way), inspect the dual-track profile (what you declared vs what your behavior suggests), and enable/disable question tuning. Conversational interface — no CLI syntax required.

    Also in 9 projects that use this

  • Saqlain-811Skill

    ios-clean

    Remove the DebugBridge SPM package and all #if DEBUG wiring from an iOS app. (gstack)

    Also in 5 projects that use this

  • Saqlain-811Skill

    ios-fix

    Autonomous iOS bug fixer. (gstack)

    Also in 5 projects that use this

  • ShreyRaviSkill

    landing-report

    Read-only queue dashboard for workspace-aware ship. Shows which VERSION slots are currently claimed by open PRs, which sibling Conductor workspaces have WIP work likely to ship soon, and what slot /ship would pick next. No mutations — just a snapshot. Use when asked to "landing report", "what's in the queue", "show me open PRs", or "which version do I claim next". (gstack)

    Also in 8 projects that use this

  • seek888Skill

    plan-devex-review

    Interactive developer experience plan review. Explores developer personas, benchmarks against competitors, designs magical moments, and traces friction points before scoring. Three modes: DX EXPANSION (competitive advantage), DX POLISH (bulletproof every touchpoint), DX TRIAGE (critical gaps only). Use when asked to "DX review", "developer experience audit", "devex review", or "API design review". Proactively suggest when the user has a plan for developer-facing products (APIs, CLIs, SDKs, libraries, platforms, docs). (gstack) Voice triggers (speech-to-text aliases): "dx review", "developer experience review", "devex review", "devex audit", "API design review", "onboarding review".

    Also in 14 projects that use this

  • timurgaleevSkill

    document-generate

    Generate missing docs for a feature, module or project as tutorials, how-tos, reference and explanation.

    Also in 7 projects that use this

  • +1 more from Saqlain-811
  • +2 more from ShreyRavi

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 38 edited versions we read keep these.

  • 6

    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:

    Read the supplied plan and relevant project instructions, existing design references, and components.

    line 13
  2. Distinctive:

    If the plan has no user-facing UI or interaction scope, explain that this design review is not applicable.

    line 19
  3. Distinctive:

    Use these dimensions together, not as seven approval gates.

    line 42
  4. Distinctive:

    When a journey is unclear, a short storyboard can expose the gap:

    line 72
  5. Distinctive:

    Do not treat silence as approval for an unresolved consequential choice.

    line 88
  6. Distinctive:

    Keep out-of-scope design ideas separate from required fixes.

    line 90

“Common” is measured: kept by at least 6 in 10 of the 38 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

Plan Design Review

Find missing user-facing design decisions before implementation. Review the plan against its audience, existing product patterns, and accepted scope.

Scope and inputs

Read the supplied plan and relevant project instructions, existing design references, and components. Use version-control context only when it resolves what changed; no PR or remote lookup is required for a local plan review. A named design-system file is useful when present, but its absence is not automatically a defect.

If the plan has no user-facing UI or interaction scope, explain that this design review is not applicable. Do not start implementation or edit product code. A review-only request remains read-only. If the user asks to improve the plan, apply clear reversible improvements directly to that plan.

Preserve deliberate design choices. Avoid inventing a new aesthetic or product requirement merely to make a review more opinionated.

Design lens

Describe what the user sees, can do, and understands, including recovery from failure. Prioritize hierarchy, interaction states, user journey, and alignment with the product's existing vocabulary. Prefer removing a distracting element over adding another. Empty states should explain the situation and offer an appropriate next action; decorative warmth is not required in every context.

Simulate relevant constraints: first use, repeat use, poor connectivity, long content, touch, keyboard, and assistive technology. Consider only cases that apply to this product. Tie taste judgments to an observable design problem and its effect on the user.

Seven-dimension rubric

Use these dimensions together, not as seven approval gates. When scoring is requested or useful for comparison, rate applicable dimensions from 0–10 with a short reason and the concrete missing specification. Scores are diagnostic; do not loop until every score reaches 10 or use an arbitrary average as proof of readiness.

DimensionReview question
Information architectureIs the primary, secondary, and tertiary content clear, with sensible navigation and grouping?
Interaction statesAre loading, empty, error, success, and partial states defined for relevant features?
User journeyDoes the plan connect entry, action, feedback, recovery, and return use?
Design specificityAre design choices concrete and suited to this product, rather than vague template language?
Design system alignmentDoes the plan reuse appropriate components, tokens, and established conventions?
Responsive and accessibilityAre meaningful viewport changes, keyboard/focus behavior, semantics, contrast, and touch behavior specified?
Unresolved decisionsWhich remaining choices would materially change implementation or the user's experience?

Generic card grids or hero sections are not defects on their own. Flag a pattern when it obscures hierarchy, invents unnecessary UI, or conflicts with the product. Replace vague phrases such as “clean, modern” with a specific decision supported by the reference or requested direction.

For interaction-heavy plans, build or repair a state table:

FeatureLoadingEmptyErrorSuccessPartial
Relevant featureVisible feedbackExplanation and next actionMessage and recoveryConfirmation and next stateWhat remains usable

Mark states not applicable with a reason. Specify triggers, what persists, focus changes, and retry/cancel behavior where they affect the interaction.

When a journey is unclear, a short storyboard can expose the gap:

StepUser actionExpected feedbackRecovery or next action
Relevant stepWhat the user doesWhat they see and understandWhat happens next

Apply findings and resolve choices

Separate clear fixes from consequential choices. For authorized plan edits, resolve obvious gaps from existing patterns without asking about each issue. For review-only work, present those same improvements as recommendations.

Batch unresolved consequential choices into one concise request, with a recommendation, alternatives, and the user-visible tradeoff for each. Continue independent review or authorized improvements while input is pending. Do not pause to ask whether to perform the review the user already requested. Do not treat silence as approval for an unresolved consequential choice.

Keep out-of-scope design ideas separate from required fixes. Add TODOs only when requested or already part of the project's authorized planning workflow; do not open a separate approval question for every potential TODO.

Acceptance and output

Completion means the requested review covers applicable dimensions, findings are concrete, and implementation-affecting unknowns are resolved or explicitly identified. For plan improvement, the authorized changes are in the plan and consistent with its acceptance criteria.

Deliver the plan path if edited, the most material findings or changes, useful scores if used, existing patterns to reuse, and any unresolved decision with its consequence. State remaining readiness gaps honestly. Do not substitute a large review report for the requested plan improvements.

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.