critique

Name: critique
Author: mphinance

15stars3forks

How to install critique?

mphinance/alpha-skills/critique

$npx -y skills add mphinance/alpha-skills --skill critique

Installs into the current project.

Summary

Run a 5-dimension expert design review on any HTML artifact in the project — Philosophy / Visual hierarchy / Detail / Functionality / Innovation, each scored 0–10. Outputs a single self-contained HTML report with a radar chart, evidence-backed scores, and three lists: Keep / Fix

Files · 1

View on GitHub

SKILL.md

1	# Critique Skill · 5 维度专家评审
2
3	Produce a single-file HTML "design review report" that scores any
4	artifact across 5 dimensions and proposes actionable fixes. Inspired by
5	the huashu-design expert-critique flow.
6
7	## When to use
8
9	- After the agent (or user) generates an artifact (deck / prototype /
10	landing page) and the user asks "what's wrong with this?" or
11	"review this"
12	- As a self-check loop the agent can run on its own output before
13	emitting it
14	- For comparing two variants of the same design
15
16	## What you produce
17
18	A single self-contained `<artifact type="text/html">` review report
19	including:
20
21	1. Header — what artifact was reviewed, date, reviewer ("OD ·
22	Critique skill"), 1-line verdict
23	2. Radar chart (inline SVG, no library) showing the 5 scores
24	3. Five dimension cards, each with:
25	- Score 0–10 (with band: 0–4 Broken · 5–6 Functional · 7–8 Strong
26	· 9–10 Exceptional)
27	- 1-paragraph evidence (cite specific elements / files / lines)
28	- One Keep / Fix / Quick-win bullet
29	4. Combined action lists at the bottom:
30	- Keep — what's working, don't touch
31	- Fix — P0 / P1 issues that are visually expensive
32	- Quick wins — 5–15 minute tweaks with disproportionate impact
33
34	## The 5 dimensions
35
36	> Each dimension is independent — a deck can be 9/10 on Innovation but
37	> 4/10 on Hierarchy and the report should say so plainly. Don't average
38	> away interesting failures.
39
40	### 1. Philosophy consistency · 哲学一致性
41
42	> Does the artifact pick a clear direction and stick to it through
43	> every micro-decision (chrome / kicker / spacing / accent)?
44
45	Evidence to look for:
46	- Is there one declared design direction (e.g. Monocle / WIRED /
47	Kinfolk) or is it three styles in a trench coat?
48	- Does the chrome / kicker vocabulary stay in one register, or does
49	page 3 say "Vol.04 · Spring" and page 7 say "BUT WAIT 🔥"?
50	- Are accent / serif / mono used by the same rule throughout?
51
52	0–4 Three styles fighting each other. 5–6 One direction but
53	half the elements drift. 7–8 Coherent, occasional drift on edge
54	pages. 9–10 Every element argues for the same thesis.
55
56	### 2. Visual hierarchy · 视觉层级
57
58	> Can a stranger figure out what to read first, second, third — without
59	> being told?
60
61	Evidence to look for:
62	- Is the largest type clearly the most important thing on each page?
63	- Do mono / serif / sans roles match the information's role (meta /
64	body / display)?
65	- Lots of "loud" elements competing? Or a clear primary + secondary +
66	tertiary tier?
67
68	0–4 Everything shouts. 5–6 Hierarchy works on hero pages but
69	breaks on body. 7–8 Clear tiers, occasional collision. 9–10 Eye
70	moves with zero friction.
71
72	### 3. Detail execution · 细节执行
73
74	> The 90/10 stuff — alignment, leading, kerning at large sizes, image
75	> framing, foot/chrome polish, edge-case spacing.
76
77	Evidence to look for:
78	- Big-stat pages: does the number sit on a baseline, or float?
79	- Left/right column tops aligned in `grid-2-7-5`?
80	- `frame-img` + caption proportions consistent across pages?
81	- Mono labels: same letter-spacing? same uppercase rule?
82	- Any orphaned `<br>` causing 1-character lines?
83
84	0–4 Visible tape and string. 5–6 Most pages clean, 1–2
85	ragged. 7–8 Polished, expert eye finds 2–3 misses. 9–10
86	Magazine-grade — the kind of detail that makes printed-by-hand
87	typographers nod.
88
89	### 4. Functionality · 功能性
90
91	> Does the artifact work for its intended use? Click targets, nav,
92	> readability at presentation distance, copy-paste-ability for code
93	> blocks, mobile fallback if relevant.
94
95	Evidence to look for:
96	- Deck: keyboard / wheel / touch nav all working? Iframe scroll
97	fallback?
98	- Landing: CTA above the fold? Phone number tappable on mobile?
99	- Runbook: code blocks copyable, mono font, no smart quotes?
100	- Critical info readable from 4m away (large screen presentation)?
101
102	0–4 Visually fine but doesn't accomplish its job. 5–6 Core
103	flow works, edge

People who liked this also viewed

6 picks

Type

critique

15stars3forks

View on GitHub

How to install critique?

mphinance/alpha-skills/critique

$npx -y skills add mphinance/alpha-skills --skill critique

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Critique Skill · 5 维度专家评审
2
3	Produce a single-file HTML "design review report" that scores any
4	artifact across 5 dimensions and proposes actionable fixes. Inspired by
5	the huashu-design expert-critique flow.
6
7	## When to use
8
9	- After the agent (or user) generates an artifact (deck / prototype /
10	landing page) and the user asks "what's wrong with this?" or
11	"review this"
12	- As a self-check loop the agent can run on its own output before
13	emitting it
14	- For comparing two variants of the same design
15
16	## What you produce
17
18	A single self-contained `<artifact type="text/html">` review report
19	including:
20
21	1. Header — what artifact was reviewed, date, reviewer ("OD ·
22	Critique skill"), 1-line verdict
23	2. Radar chart (inline SVG, no library) showing the 5 scores
24	3. Five dimension cards, each with:
25	- Score 0–10 (with band: 0–4 Broken · 5–6 Functional · 7–8 Strong
26	· 9–10 Exceptional)
27	- 1-paragraph evidence (cite specific elements / files / lines)
28	- One Keep / Fix / Quick-win bullet
29	4. Combined action lists at the bottom:
30	- Keep — what's working, don't touch
31	- Fix — P0 / P1 issues that are visually expensive
32	- Quick wins — 5–15 minute tweaks with disproportionate impact
33
34	## The 5 dimensions
35
36	> Each dimension is independent — a deck can be 9/10 on Innovation but
37	> 4/10 on Hierarchy and the report should say so plainly. Don't average
38	> away interesting failures.
39
40	### 1. Philosophy consistency · 哲学一致性
41
42	> Does the artifact pick a clear direction and stick to it through
43	> every micro-decision (chrome / kicker / spacing / accent)?
44
45	Evidence to look for:
46	- Is there one declared design direction (e.g. Monocle / WIRED /
47	Kinfolk) or is it three styles in a trench coat?
48	- Does the chrome / kicker vocabulary stay in one register, or does
49	page 3 say "Vol.04 · Spring" and page 7 say "BUT WAIT 🔥"?
50	- Are accent / serif / mono used by the same rule throughout?
51
52	0–4 Three styles fighting each other. 5–6 One direction but
53	half the elements drift. 7–8 Coherent, occasional drift on edge
54	pages. 9–10 Every element argues for the same thesis.
55
56	### 2. Visual hierarchy · 视觉层级
57
58	> Can a stranger figure out what to read first, second, third — without
59	> being told?
60
61	Evidence to look for:
62	- Is the largest type clearly the most important thing on each page?
63	- Do mono / serif / sans roles match the information's role (meta /
64	body / display)?
65	- Lots of "loud" elements competing? Or a clear primary + secondary +
66	tertiary tier?
67
68	0–4 Everything shouts. 5–6 Hierarchy works on hero pages but
69	breaks on body. 7–8 Clear tiers, occasional collision. 9–10 Eye
70	moves with zero friction.
71
72	### 3. Detail execution · 细节执行
73
74	> The 90/10 stuff — alignment, leading, kerning at large sizes, image
75	> framing, foot/chrome polish, edge-case spacing.
76
77	Evidence to look for:
78	- Big-stat pages: does the number sit on a baseline, or float?
79	- Left/right column tops aligned in `grid-2-7-5`?
80	- `frame-img` + caption proportions consistent across pages?
81	- Mono labels: same letter-spacing? same uppercase rule?
82	- Any orphaned `<br>` causing 1-character lines?
83
84	0–4 Visible tape and string. 5–6 Most pages clean, 1–2
85	ragged. 7–8 Polished, expert eye finds 2–3 misses. 9–10
86	Magazine-grade — the kind of detail that makes printed-by-hand
87	typographers nod.
88
89	### 4. Functionality · 功能性
90
91	> Does the artifact work for its intended use? Click targets, nav,
92	> readability at presentation distance, copy-paste-ability for code
93	> blocks, mobile fallback if relevant.
94
95	Evidence to look for:
96	- Deck: keyboard / wheel / touch nav all working? Iframe scroll
97	fallback?
98	- Landing: CTA above the fold? Phone number tappable on mobile?
99	- Runbook: code blocks copyable, mono font, no smart quotes?
100	- Critical info readable from 4m away (large screen presentation)?
101
102	0–4 Visually fine but doesn't accomplish its job. 5–6 Core
103	flow works, edge

People who liked this also viewed

6 picks

Type