design-review
Skill19 starsChanged 22 days agoApache-2.0View on GitHub
Use when the user asks to review, assess, or check a lens design against a spec. Also use after optimization to evaluate results. Triggers on "how does this look", "check the design", "review", "evaluate".
167 other projects use a version of this.
Adopted by
Most recently taken up by
- gspencergoog/skillstheir own version, first seen Nov 16, 2026
- schen22/claude-configtheir own version, first seen Nov 16, 2026
- seancabalse/agentic-playbooktheir own version, first seen Nov 16, 2026
- Taro000/req-to-designtheir own version, first seen Nov 15, 2026
- persiyanov/skillstheir own version, first seen Nov 15, 2026
- SergioRibera/definitive-uxtheir own version, first seen Nov 15, 2026
Used with
Projects with it also use team-level, 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.
CoralGameSkillteam-level
Orchestrate level design team: level-designer + narrative-director + world-builder + art-director + systems-designer + qa-tester for complete area/level creation.
Also in 13 projects that use this
CoralGameSkillteam-audio
Orchestrate audio team: audio-director + sound-designer + technical-artist + gameplay-programmer for full audio pipeline from direction to implementation.
Also in 14 projects that use this
lilangMaxSkillasset-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 11 projects that use this
lilangMaxSkillcreate-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 11 projects that use this
julianoczkowskiSkillbrief-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
- +4 more from lilangMax
- +3 more from CoralGame
Other versions167 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 167
- garrytan/gstack★ 142ksame name
- Donchitos/Claude-Code-Game-Studios★ 39ksame name
- plugin87/ux-ui-agent-skills★ 3.2ksame name
- DenchHQ/DenchClaw★ 1.7ksame name
- DenchHQ/ironclaw★ 1.7ksame name
- me2resh/apexyard★ 542same name
- julianoczkowski/designer-skills★ 536same name
- openclaw/openclaw-enterprise★ 472same name
- AndrewDryga/emisar★ 376same name
- TheDecipherist/claude-code-mastery-project-starter-kit★ 337same name
157 more not listed
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 repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- bench-catalogSkill
- design-aberration-reviewSkill
- design-compareSkill
- design-glass-matchingSkill
- design-tolerance-desensitizeSkill
- design-vision-reviewSkill
- optimize-loopSkill
- zos-api-debugSkill
The file
We split this skill into the rules it gives an agent
10 rules. 0 are common, 8 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 139 edited versions we read keep these.
8
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
- Distinctive: line 17
After loading a design ("how does this look?")
- Distinctive: line 18
After optimization ("did it improve?")
- Distinctive: line 19
Before saving ("is this worth keeping?")
- Distinctive: line 20
When the user asks to "review", "check", "assess", or "evaluate"
- Very specific: line 21
NOT for: comparing multiple designs (use /design-compare)
- Distinctive: line 27
If the user provided a spec (EFL, F-number, max spot size, MTF target, field angles, wavelength band), use those.
- Distinctive: line 31
If the user gave a *partial* spec — first-order targets but no image-quality bands, say — do not silently invent the missing ones.
- Distinctive: line 47
Run steps 1-2 first (fast prescription reads), then 3-5 (heavier analyses).
- Very specific: line 90
Keep every section. If a section has nothing in it, say why rather than dropping it — an absent Manufacturability block reads as "clean" …
- Distinctive: line 99
Keep suggestions brief. The user is an optical engineer — they need the data organized, not a lecture.
“Common” is measured: kept by at least 6 in 10 of the 139 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.
Files
Design Review — Comprehensive Lens Assessment
Overview
One-command lens assessment. Pulls the relevant analysis metrics through the ZOS-API, compares them against the user's spec, and outputs a structured report. Manufacturability ratios are surfaced as prompted feedback to the user — the user brings the acceptance bands; this skill does not bake a verdict.
When to Use
- After loading a design ("how does this look?")
- After optimization ("did it improve?")
- Before saving ("is this worth keeping?")
- When the user asks to "review", "check", "assess", or "evaluate"
- NOT for: comparing multiple designs (use
/design-compare)
Workflow
Step 1: Gather spec targets
If the user provided a spec (EFL, F-number, max spot size, MTF target, field angles, wavelength band), use those. If not, ask: "What are your targets? (or I can just report current state)".
If the user gave a partial spec — first-order targets but no image-quality bands, say — do not silently invent the missing ones. Report the un-targeted metrics un-graded, and say in the report which columns have no target behind them.
Step 2: Run the analysis sequence
Pull these through the ZOS-API typed analyses, in order:
- System info — EFL, F-number, total track length, surface count, wavelengths, field set.
- Prescription — surface-by-surface radii, thicknesses, glasses.
- Spot — RMS and GEO spot size for ALL fields.
- MTF — MTF curves for ALL fields (report at two representative frequencies, e.g. 50 and 100 lp/mm).
- Wavefront — on-axis wavefront error (RMS waves) on a flat image plane.
Run steps 1-2 first (fast prescription reads), then 3-5 (heavier analyses).
get_mtf may return one more data series than you have fields — the extra one
is the diffraction limit, not a field. Check the series count against the field
count before you map them, and label the limit as the limit.
Step 3: Output report
## Design Review
### System Summary
| Parameter | Value | Target | Status |
|--------------|-------|--------|--------|
| EFL | ... | ... | PASS/FAIL |
| F-number | ... | ... | PASS/FAIL |
| Total track | ... | ... | — |
| Elements | ... | — | — |
### Image Quality
| Field | RMS Spot (um) | Target | MTF@50 | MTF@100 | Status |
|------------|---------------|--------|--------|---------|--------|
| On-axis | ... | ... | ... | ... | PASS/FAIL |
| 0.7 field | ... | ... | ... | ... | PASS/FAIL |
| Full field | ... | ... | ... | ... | PASS/FAIL |
### Wavefront
- On-axis RMS: ... waves (flat image plane)
- Diffraction limited: YES/NO
### Manufacturability (read-only ratios — bring your own bands)
- Min center / edge thickness and aspect ratios as raw numbers (from `check_clearance`).
- No baked verdict — these are prompts for the user's judgement.
### Verdict: [PASS / NEEDS WORK / FAIL]
[One sentence: what's good, what needs attention.]The report ends with the verdict heading. It begins with the literal token
### Verdict: followed by exactly one of PASS, NEEDS WORK, or FAIL. Keep that
marker exactly as written, so the verdict stays easy to find for the reader and in later
sessions — do not substitute prose for it ("Bottom line", "Summary", "Overall"). Anything else you want to say goes above the line.
Keep every section. If a section has nothing in it, say why rather than dropping it
— an absent Manufacturability block reads as "clean" when it means "not measured."
Step 4: Suggest next steps (only if NEEDS WORK or FAIL)
- Spot too large: "Consider freeing more radii or adding an element."
- MTF low at edge: "Field curvature may be the limiter — check the wavefront."
- EFL off target: "Add an EFL operand to the merit function."
Keep suggestions brief. The user is an optical engineer — they need the data organized, not a lecture.
Key Rules
- Evaluate the wavefront on a flat image plane unless the user explicitly asks for a best-focus refocus. If you also report a best-focus figure, label it as a refocus and keep the flat-plane number as the headline — a best-focus RMS can be 2× rosier, and quoting it unqualified overstates the design.
- Surface ratios are prompts, not verdicts. The user owns the acceptance bands.
- Report the worst field, always. A summary sentence about field performance must be supported by the corner row, not by the fields that flatter the design.
- Ground any operand or glass you are unsure about with the reference tools
(
search_reference/lookup_operand) — do not guess a code.
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
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.

