brand-kit

Name: brand-kit
Author: matteotitta

36stars14forks

How to install brand-kit?

matteotitta/claude-code-marketing-quickstart/brand-kit

$npx -y skills add matteotitta/claude-code-marketing-quickstart --skill brand-kit

Installs into the current project.

Summary

Extract visual identity from screenshots (primary) or website URL (fallback) and compile into a DESIGN.md-format brand kit — YAML token frontmatter (colors, typography, rounded, spacing, components) + 8 ordered prose sections. The YAML tokens are machine-authoritative; downstream

Files · 1

View on GitHub

SKILL.md

1	# brand-kit — Stage 1 + 2 research skill (PMM spine)
2
3	Compiles visual identity into a structured brand kit. Output format: DESIGN.md (YAML token frontmatter + 8 prose sections). The YAML tokens are the machine-authoritative source — downstream skills that produce visual output (landing-page wireframes, ad creative briefs, slide decks) consume the tokens directly.
4
5	---
6
7	## When to use
8
9	- When the team refreshes visual identity (new color palette, new typography system)
10	- Before building any landing page, ad creative, or designed artifact — they need the brand kit as input
11	- Quarterly refresh
12
13	## When NOT to use
14
15	- For voice / tone (use `/tov-guidelines` — voice + visual are separate but both live in `marketing/brand/`)
16	- For positioning or messaging (use `/positioning` or `/product-messaging`)
17	- For one-off design checks (write a custom gate skill that reads `brand-kit.md` tokens)
18
19	## How it works
20
21	1. Inputs: screenshots of your product / website (primary, higher fidelity) OR website URL (fallback, lower fidelity).
22	2. Reads: visual content via Claude's vision (screenshots) or Exa MCP web fetch (URL).
23	3. Extracts: exact colors (hex codes), typography stack (font family + sizes + weights + line heights + letter spacing), spacing scale, corner radii, component patterns (button states, card styles, input styles, etc.).
24	4. Produces a DESIGN.md-format brand kit:
25	- YAML token frontmatter (machine-authoritative): colors, typography, rounded, spacing, components — token references use `{path.to.token}` syntax
26	- 8 ordered prose sections: Overview, Colors, Typography, Layout, Elevation & Depth, Shapes, Components, Do's and Don'ts
27	5. Validates against the design-production doctrine: ≤2 brand colors active per screen, ≤2 font weights per screen, ≤10% accent coverage, no banned patterns (gradient text, side-stripe borders, hero-metric template, icon-tile-above-heading template, cardocalypse, glassmorphism).
28	6. Writes to `marketing/brand/brand-kit.md` (overwrites prior canonical; git history preserves prior versions).
29
30	## Invoke
31
32	```
33	/brand-kit screenshots: [paths to product screenshots]
34	```
35
36	Or URL-fallback:
37	```
38	/brand-kit https://yourdomain.com — extract visual identity
39	```
40
41	## Example output
42
43	See [`marketing/brand/brand-kit.md`](../../../marketing/brand/brand-kit.md) for the PulseAnalytics seed. Notice the YAML frontmatter is the source of truth; the prose body explains intent. Token references use `{path.to.token}` syntax — every component reference resolves to a defined token. Do's and Don'ts encode quantitative limits (≤2 font weights per screen, ≤10% accent coverage).
44
45	## Dependencies
46
47	- Reads from: source visual content (screenshots or URL — provided as input)
48	- Reads via Exa MCP (optional): website if URL-fallback path is used
49	- Writes to: `marketing/brand/brand-kit.md` (canonical)
50	- Downstream readers: `/landing-page-wireframe`, `/landing-page-copy`, `/ad-creative-brief`, `/sales-deck`, `/linkedin-carousels`, any visual-output skill
51
52	## Customization
53
54	Add custom token categories if your brand requires them (e.g., `motion` tokens for animation timing, `iconography` tokens for icon style). The YAML frontmatter accepts arbitrary top-level keys; downstream skills consume whatever exists.
55
56	Update the Do's and Don'ts when you spot a recurring violation in produced artifacts — add a rule, downstream skills enforce it on next run.
57
58	## Where this fits in the Example 1 chain
59
60	```
61	Week 1: /brand-kit (parallel with positioning + voice research) → visual identity locked
62	Week 2: /landing-page-wireframe + /landing-page-copy read brand kit + messaging
63	Week 3+: every visual-output skill reads brand kit
64	```
65
66	## Refresh cadence
67
68	Quarterly. Trigger sooner on visual identity refresh, new product launch with distinct visual treatment, or major

People who liked this also viewed

6 picks

Type

brand-kit

36stars14forks

View on GitHub

How to install brand-kit?

matteotitta/claude-code-marketing-quickstart/brand-kit

$npx -y skills add matteotitta/claude-code-marketing-quickstart --skill brand-kit

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# brand-kit — Stage 1 + 2 research skill (PMM spine)
2
3	Compiles visual identity into a structured brand kit. Output format: DESIGN.md (YAML token frontmatter + 8 prose sections). The YAML tokens are the machine-authoritative source — downstream skills that produce visual output (landing-page wireframes, ad creative briefs, slide decks) consume the tokens directly.
4
5	---
6
7	## When to use
8
9	- When the team refreshes visual identity (new color palette, new typography system)
10	- Before building any landing page, ad creative, or designed artifact — they need the brand kit as input
11	- Quarterly refresh
12
13	## When NOT to use
14
15	- For voice / tone (use `/tov-guidelines` — voice + visual are separate but both live in `marketing/brand/`)
16	- For positioning or messaging (use `/positioning` or `/product-messaging`)
17	- For one-off design checks (write a custom gate skill that reads `brand-kit.md` tokens)
18
19	## How it works
20
21	1. Inputs: screenshots of your product / website (primary, higher fidelity) OR website URL (fallback, lower fidelity).
22	2. Reads: visual content via Claude's vision (screenshots) or Exa MCP web fetch (URL).
23	3. Extracts: exact colors (hex codes), typography stack (font family + sizes + weights + line heights + letter spacing), spacing scale, corner radii, component patterns (button states, card styles, input styles, etc.).
24	4. Produces a DESIGN.md-format brand kit:
25	- YAML token frontmatter (machine-authoritative): colors, typography, rounded, spacing, components — token references use `{path.to.token}` syntax
26	- 8 ordered prose sections: Overview, Colors, Typography, Layout, Elevation & Depth, Shapes, Components, Do's and Don'ts
27	5. Validates against the design-production doctrine: ≤2 brand colors active per screen, ≤2 font weights per screen, ≤10% accent coverage, no banned patterns (gradient text, side-stripe borders, hero-metric template, icon-tile-above-heading template, cardocalypse, glassmorphism).
28	6. Writes to `marketing/brand/brand-kit.md` (overwrites prior canonical; git history preserves prior versions).
29
30	## Invoke
31
32	```
33	/brand-kit screenshots: [paths to product screenshots]
34	```
35
36	Or URL-fallback:
37	```
38	/brand-kit https://yourdomain.com — extract visual identity
39	```
40
41	## Example output
42
43	See [`marketing/brand/brand-kit.md`](../../../marketing/brand/brand-kit.md) for the PulseAnalytics seed. Notice the YAML frontmatter is the source of truth; the prose body explains intent. Token references use `{path.to.token}` syntax — every component reference resolves to a defined token. Do's and Don'ts encode quantitative limits (≤2 font weights per screen, ≤10% accent coverage).
44
45	## Dependencies
46
47	- Reads from: source visual content (screenshots or URL — provided as input)
48	- Reads via Exa MCP (optional): website if URL-fallback path is used
49	- Writes to: `marketing/brand/brand-kit.md` (canonical)
50	- Downstream readers: `/landing-page-wireframe`, `/landing-page-copy`, `/ad-creative-brief`, `/sales-deck`, `/linkedin-carousels`, any visual-output skill
51
52	## Customization
53
54	Add custom token categories if your brand requires them (e.g., `motion` tokens for animation timing, `iconography` tokens for icon style). The YAML frontmatter accepts arbitrary top-level keys; downstream skills consume whatever exists.
55
56	Update the Do's and Don'ts when you spot a recurring violation in produced artifacts — add a rule, downstream skills enforce it on next run.
57
58	## Where this fits in the Example 1 chain
59
60	```
61	Week 1: /brand-kit (parallel with positioning + voice research) → visual identity locked
62	Week 2: /landing-page-wireframe + /landing-page-copy read brand kit + messaging
63	Week 3+: every visual-output skill reads brand kit
64	```
65
66	## Refresh cadence
67
68	Quarterly. Trigger sooner on visual identity refresh, new product launch with distinct visual treatment, or major

People who liked this also viewed

6 picks

Type