agentleFS
grace-methodgrace-method/grace-plugin/skills/design-review/SKILL.md

design-review

Skill0 starsChanged 2 months agoMITView on GitHub

Design and architecture review. Walk the operator's procedures, then decompose approved business-analysis solutions into components, interfaces, ADRs, and technology choices. Use after business analysis is approved and before implementation begins.

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

9 rules. 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 129 edited versions we read keep these.

  • 9

    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:

    Use this skill to turn approved Stage 2 analysis into a reviewable Stage 3 design package.

    line 14
  2. Distinctive:

    Check each step against the state the previous step actually leaves behind, not the state the design assumes.

    line 30
  3. Distinctive:

    every component should trace to one or more solutions

    line 53
  4. Distinctive:

    every mapped component should also trace to requirement(s)

    line 54
  5. Distinctive:

    call out solutions that do not yet have an implementing design element

    line 57
  6. Distinctive:

    call out components or interfaces that do not trace back cleanly

    line 58
  7. Distinctive:

    do not treat design as settled until a human reviews it

    line 61
  8. Distinctive:

    do not move into implementation until the design package is approved enough

    line 62
  9. Distinctive:

    Use the templates in this skill directory when scaffolding or substantially reshaping those artifacts.

    line 84

“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

Use this skill to turn approved Stage 2 analysis into a reviewable Stage 3 design package.

Approved Business Analysis Solutions

!cat docs/02-analysis/business-analysis.md 2>/dev/null || echo "WARNING: No business analysis found. Analysis should precede design."

Requirements

!cat docs/02-analysis/requirements/requirements.md 2>/dev/null || echo "No requirements found."

Interaction Pattern

  1. Review the approved solutions from business analysis.
    • Work from the analysis and requirements rather than inventing a parallel problem framing.
  2. Walk the operator's procedures before defining components.
    • For each job in the Product Definition's task inventory, write the steps in the order a human performs them.
    • Check each step against the state the previous step actually leaves behind, not the state the design assumes. Follow the instruction and take the next step — do not reason about the code path instead.
    • Cover the whole calendar, not the representative path: first run, period-end, year-end, recovery, and re-entry after a gap. Take the rarest first; nothing rehearses them.
    • Of a rare step ask: can it be done at all? Of a frequent step ask: what does it cost, and will the friction get worked around?
    • This applies to any human-operated sequence whose steps are separated in time and inherit state from one another. Recurrence is a common cause, not the test.
    • Record the result in docs/06-operations/user-manual.md, using the template in this skill directory. The walk is the design activity; the manual is its record.
    • The manual describes what the product does now. Behaviour that is designed but has not shipped goes in exactly one block per release, marked <!-- UNSHIPPED: R-NNN --> … <!-- /UNSHIPPED: R-NNN -->, never inline. Closure removes the block or promotes its text; inline futures become claims the product does not honour, and nothing can find them later.
  3. For each solution, or for each coherent cluster of related solutions, define:
    • major components and their responsibilities
    • communication paths between components
    • key boundaries, interfaces, or contracts
    • technology choices with rationale and tradeoffs
  4. Apply Design by Contract at important boundaries.
    • identify preconditions
    • identify postconditions
    • identify important invariants where relevant
  5. Record significant decisions as ADRs when the choice has meaningful consequences.
    • include context, decision, and consequences
    • An ADR's status is a claim about whether the decision binds. When the design gate accepts one, stamp it Accepted at the <R-NNN> design gate, <YYYY-MM-DD> in the gate's own commit. Never leave Proposed behind after a gate has passed — a reader cannot then tell a settled decision from an open one.
  6. Build the design traceability matrix.
    • every component should trace to one or more solutions
    • every mapped component should also trace to requirement(s)
  7. Flag gaps.
    • call out solutions that do not yet have an implementing design element
    • call out components or interfaces that do not trace back cleanly
  8. Present the design package for approval.
    • do not treat design as settled until a human reviews it
    • do not move into implementation until the design package is approved enough

Key Questions

  • What are the major components and their responsibilities?
  • How do they communicate?
  • What are the important technical tradeoffs and why?
  • What constraints does the chosen technology impose?
  • Where are the meaningful boundaries?
  • How does a human operate this over time, including the steps that run only once a year?

Outputs

Produce or update these artifacts as needed:

  • docs/03-design/01-components.md
  • docs/03-design/02-interfaces.md
  • docs/03-design/03-technology-choices.md
  • docs/03-design/04-traceability-matrix.md
  • docs/03-design/adr/
  • docs/06-operations/user-manual.md — created here, maintained for the product's life

Use the templates in this skill directory when scaffolding or substantially reshaping those artifacts.

Gate

Human approval is required before implementation begins.

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.