get-shit-pretty 0.5.2 → 0.6.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "gsp",
3
- "version": "0.5.2",
3
+ "version": "0.6.0",
4
4
  "description": "Design engineering system for AI coding agents. Brand identity + design projects, from strategy to code.",
5
5
  "author": {
6
6
  "name": "jubscodes",
package/bin/install.js CHANGED
@@ -1704,6 +1704,9 @@ function finishInstall(settingsPath, settings, statuslineCommand, shouldInstallS
1704
1704
  if (!onboardingShown && !hasQuiet) {
1705
1705
  onboardingShown = true;
1706
1706
  console.log(`
1707
+ ${c.secondary}Design engineering for AI coding tools.${c.reset}
1708
+ ${c.secondary}Brand strategy, visual identity, design systems, UI — built by agents.${c.reset}
1709
+
1707
1710
  ${c.bold}Get started:${c.reset}
1708
1711
  ${c.accent}${newCmd}${c.reset} ${c.secondary}start here — brand, project, or both${c.reset}
1709
1712
  ${c.accent}${helpCmd}${c.reset} ${c.secondary}all commands${c.reset}
@@ -50,7 +50,7 @@ Build a single screen. You receive only that screen's design chunk and its refer
50
50
  ### `full`
51
51
  Legacy mode — build everything in one pass. Used as backward-compatible default.
52
52
 
53
- **Chunk-aware mode:** When chunks are provided instead of full monoliths (screen chunks from `design/`, brief chunks from `brief/`, research specs from `research/`, component chunks from brand `patterns/components/`), work with the focused context. Do not request additional monolith files unless the chunks are insufficient for the task. This keeps your context lean and focused on the specific screen being built.
53
+ **Chunk-aware mode:** Work with the chunk context provided. Do not request additional files unless the chunks are insufficient for the task. This keeps your context lean and focused on the specific screen being built.
54
54
 
55
55
  **Revision mode:** When `review/issues.md` is provided, you are re-entering the build phase to address QA issues. Read the issues, fix them in the codebase, and update BUILD-LOG.md with the revision.
56
56
  </role>
@@ -1,42 +1,37 @@
1
1
  ---
2
2
  name: accessibility
3
- description: Accessibility audit — contrast checks, WCAG compliance, code audits, token verification
3
+ description: Quick contrast checks and token WCAG audits inline, no agent
4
4
  user-invocable: true
5
5
  allowed-tools:
6
6
  - Read
7
7
  - Write
8
8
  - Bash
9
- - Agent
10
9
  - Glob
11
10
  - Grep
12
11
  - AskUserQuestion
13
12
  ---
14
13
  <context>
15
14
  Standalone composable accessibility skill. Works two ways:
16
- 1. **Standalone** — user runs `/gsp:accessibility` directly for design, code, or token audits
15
+ 1. **Standalone** — user runs `/gsp:accessibility` directly for quick contrast checks or token audits
17
16
  2. **As a building block** — critique and review phases detect prior accessibility output and reuse it
18
17
 
19
18
  Follows the composable pattern: deterministic modes, predictable output paths, filesystem as integration layer.
19
+
20
+ For full design audits, code audits, or statement generation, use `/gsp:accessibility-audit`.
20
21
  </context>
21
22
 
22
23
  <objective>
23
- Run accessibility audits in multiple modes design, code, tokens, quick check, or statement generation.
24
+ Run lightweight accessibility checks inlinecontrast ratio lookups and token WCAG verification.
24
25
 
25
26
  **Input:** Mode flag + optional arguments
26
- **Output:** Audit chunks in the appropriate project directory
27
- **Agent:** `gsp-accessibility-auditor` (for design and code modes), inline for tokens/check/statement
27
+ **Output:** Display output (check mode) or audit chunk (token mode)
28
+ **Agent:** None this skill runs entirely inline
28
29
  </objective>
29
30
 
30
- <execution_context>
31
- @${CLAUDE_SKILL_DIR}/../../prompts/08-accessibility-auditor.md
32
- @${CLAUDE_SKILL_DIR}/../../references/wcag-checklist.md
33
- </execution_context>
34
-
35
31
  <rules>
36
32
  - Always use `AskUserQuestion` for user interaction — never prompt via plain text
37
33
  - Quick check mode (`--check`) produces display output only — no files written
38
34
  - Token audit mode runs inline — no agent spawned
39
- - Statement mode reads prior audit results — fails gracefully if none exist
40
35
  - Default conformance level is AA unless overridden by `--level AAA` or config
41
36
  - Foundation chunks follow `references/chunk-format.md` format
42
37
  </rules>
@@ -46,39 +41,34 @@ Run accessibility audits in multiple modes — design, code, tokens, quick check
46
41
 
47
42
  Read `$ARGUMENTS` to determine the mode:
48
43
 
49
- | Input | Mode | Agent? | Output |
50
- |-------|------|--------|--------|
51
- | (no args) | Design audit on `.design/` chunks | Yes (`gsp-accessibility-auditor`) | `critique/accessibility-audit.md` + `accessibility-fixes.md` |
52
- | `--tokens` | Token-only: contrast pairs, sizing, spacing | No (inline) | `critique/accessibility-token-audit.md` |
53
- | `--code` | Codebase audit: ARIA, keyboard, semantic HTML | Yes (`gsp-accessibility-auditor`) | `review/accessibility-audit.md` + `accessibility-fixes.md` |
54
- | `--statement` | Generate accessibility statement from prior audits | No (inline) | `exports/accessibility-statement.md` |
55
- | `--check #FG #BG` | Quick contrast check | No (inline, no files) | Display only |
44
+ | Input | Mode | Output |
45
+ |-------|------|--------|
46
+ | `--check #FG #BG` | Quick contrast check | Display only |
47
+ | `--tokens` | Token-only: contrast pairs, sizing, spacing | `critique/accessibility-token-audit.md` |
48
+ | (no args) | Mode picker | Prompt user |
56
49
 
57
50
  Additional flag: `--level AAA` overrides conformance level (default: AA).
58
51
 
59
- ## Step 2: Resolve context
52
+ ## Step 2: Route by mode
60
53
 
61
- ### Quick check mode (`--check`)
54
+ ### No args mode picker
62
55
 
63
- If args contain `--check`, extract the two hex color values and skip to Step 3.
56
+ If no arguments provided, use `AskUserQuestion`:
64
57
 
65
- ### All other modes
58
+ **"What would you like to do?"**
59
+ - **Quick contrast check** — "check specific color pairs for WCAG contrast compliance"
60
+ - **Token audit** — "audit tokens.json for WCAG compliance"
61
+ - **Full design/code audit** — "run `/gsp:accessibility-audit` for full WCAG audits, code audits, or statement generation"
66
62
 
67
- Scan `.design/projects/` for project directories. If only one project exists, use it. If multiple, use `AskUserQuestion` to ask which project.
63
+ If user picks "Full design/code audit", tell them to run `/gsp:accessibility-audit` and stop.
68
64
 
69
- Set `PROJECT_PATH` = `.design/projects/{project}`
65
+ ### Quick check mode (`--check`)
70
66
 
71
- Read `{PROJECT_PATH}/config.json` to get:
72
- - `accessibility_level` — override conformance level (if not set via `--level` flag)
73
- - `implementation_target` — needed for code mode
67
+ If args contain `--check`, extract the two hex color values and skip to Step 3.
74
68
 
75
- Read `{PROJECT_PATH}/brand.ref` to resolve brand path:
76
- - Set `BRAND_PATH` = `.design/branding/{brand}`
69
+ ### Token audit mode (`--tokens`)
77
70
 
78
- Determine final conformance level:
79
- 1. `--level` flag (highest priority)
80
- 2. `accessibility_level` from config.json
81
- 3. Default: "WCAG 2.2 AA"
71
+ Skip to Step 4.
82
72
 
83
73
  ## Step 3: Quick check mode (`--check #FG #BG`)
84
74
 
@@ -114,7 +104,24 @@ Convert hex to relative luminance (sRGB linearization), then:
114
104
 
115
105
  ## Step 4: Token audit mode (`--tokens`)
116
106
 
117
- Read token and palette files from the brand/project:
107
+ ### Resolve context
108
+
109
+ Resolve project from `.design/projects/` (one → use it, multiple → ask). Set `PROJECT_PATH`.
110
+
111
+ Read `{PROJECT_PATH}/config.json` to get:
112
+ - `accessibility_level` — override conformance level (if not set via `--level` flag)
113
+
114
+ Read `{PROJECT_PATH}/brand.ref` to resolve brand path:
115
+ - Set `BRAND_PATH` = `.design/branding/{brand}`
116
+
117
+ Determine final conformance level:
118
+ 1. `--level` flag (highest priority)
119
+ 2. `accessibility_level` from config.json
120
+ 3. Default: "WCAG 2.2 AA"
121
+
122
+ ### Read token and palette files
123
+
124
+ Read from the brand/project:
118
125
  - `{BRAND_PATH}/identity/palettes.json`
119
126
  - `{BRAND_PATH}/identity/color-system.md`
120
127
  - `{BRAND_PATH}/patterns/tokens.json`
@@ -124,28 +131,28 @@ If files don't exist, report which are missing and stop.
124
131
 
125
132
  ### Token checks
126
133
 
127
- **5.1 Contrast Pairs:**
134
+ **4.1 Contrast Pairs:**
128
135
  - Extract every semantic foreground/background pair from tokens.json
129
136
  - Calculate WCAG 2.x contrast ratio for each pair
130
137
  - Flag failures: normal text < 4.5:1, large text < 3:1, non-text < 3:1
131
138
 
132
- **5.2 Interactive States:**
139
+ **4.2 Interactive States:**
133
140
  - Check hover, active, focus, disabled state color pairs
134
141
  - Verify disabled states still meet 3:1 non-text contrast
135
142
 
136
- **5.3 Focus Ring:**
143
+ **4.3 Focus Ring:**
137
144
  - Find focus ring token — check >= 3:1 contrast against adjacent backgrounds
138
145
  - Verify ring width >= 2px
139
146
 
140
- **5.4 Dark Mode:**
147
+ **4.4 Dark Mode:**
141
148
  - If dark mode tokens exist, re-verify all contrast pairs
142
149
  - Dark mode is a separate verification pass, not assumed from light mode
143
150
 
144
- **5.5 Touch Targets:**
151
+ **4.5 Touch Targets:**
145
152
  - Check button/link sizing tokens >= 44px for primary actions, >= 24px minimum
146
153
  - Check spacing tokens between adjacent interactive elements
147
154
 
148
- **5.6 Typography Minimums:**
155
+ **4.6 Typography Minimums:**
149
156
  - Body text >= 16px (1rem)
150
157
  - Caption/small text >= 12px
151
158
  - Line-height >= 1.5 for body text
@@ -193,156 +200,14 @@ Conformance target: {level}
193
200
  ### Completion
194
201
 
195
202
  Display result and use `AskUserQuestion`:
196
- - **Run full design audit** — "audit design screens for WCAG compliance"
197
- - **Run code audit** — "check the codebase for accessibility issues"
203
+ - **Run full design audit** — "run `/gsp:accessibility-audit` for full WCAG design audit"
204
+ - **Run code audit** — "run `/gsp:accessibility-audit --code` to check the codebase"
198
205
  - **Done** — "that's all for now"
199
206
 
200
- ## Step 5: Design audit mode (default, no flags)
201
-
202
- Verify design chunks exist:
203
- - Read `{PROJECT_PATH}/design/INDEX.md` to find screen chunks
204
- - If no design chunks, tell user to complete design phase first and stop
205
-
206
- ### Spawn agent
207
-
208
- Spawn `gsp-accessibility-auditor` with:
209
- - All design chunks from `{PROJECT_PATH}/design/`
210
- - Brand identity context (color system, typography)
211
- - Brand system context (tokens, components)
212
- - Conformance level
213
- - WCAG checklist reference
214
- - **Output path:** `{PROJECT_PATH}/critique/`
215
- - **Instructions:** "Audit all design screens against {level}. Write `accessibility-audit.md` and `accessibility-fixes.md` to the output path."
216
-
217
- ### Completion
218
-
219
- Display result:
220
-
221
- ```
222
- /gsp:accessibility — design audit complete
223
- ═══════════════════════════════════════
224
-
225
- {PROJECT_PATH}/critique/
226
- ├── accessibility-audit.md
227
- └── accessibility-fixes.md
228
-
229
- ─────────────────────────────────────
230
- ```
231
-
232
- Use `AskUserQuestion`:
233
- - **Run token audit** — "check design token contrast pairs"
234
- - **Continue to build** — "implement designs in the codebase"
235
- - **View audit** — "read the accessibility report"
236
- - **Done** — "that's all for now"
237
-
238
- ## Step 6: Code audit mode (`--code`)
239
-
240
- Determine codebase scope:
241
- - Read `{PROJECT_PATH}/config.json` for `implementation_target`
242
- - If build phase completed, read `{PROJECT_PATH}/build/BUILD-LOG.md` for file paths
243
- - Otherwise, use `implementation_target` to determine where to look
244
-
245
- ### Spawn agent
246
-
247
- Spawn `gsp-accessibility-auditor` with:
248
- - Codebase paths to audit
249
- - Brand system tokens (for contrast verification against hardcoded values)
250
- - Conformance level
251
- - WCAG checklist reference
252
- - **Output path:** `{PROJECT_PATH}/review/`
253
- - **Instructions:** "Code audit mode. Use Grep and Glob to find accessibility issues in the codebase. Check ARIA, keyboard handlers, semantic HTML, heading hierarchy, alt text, lang attributes, skip-nav, focus management. Write `accessibility-audit.md` and `accessibility-fixes.md` to the output path with actual file paths and line numbers."
254
-
255
- ### Completion
256
-
257
- Display result:
258
-
259
- ```
260
- /gsp:accessibility --code — code audit complete
261
- ═══════════════════════════════════════
262
-
263
- {PROJECT_PATH}/review/
264
- ├── accessibility-audit.md
265
- └── accessibility-fixes.md
266
-
267
- ─────────────────────────────────────
268
- ```
269
-
270
- Use `AskUserQuestion`:
271
- - **Fix issues** — "address the accessibility issues found"
272
- - **Generate statement** — "create an accessibility statement"
273
- - **View audit** — "read the code accessibility report"
274
- - **Done** — "that's all for now"
275
-
276
- ## Step 7: Statement mode (`--statement`)
277
-
278
- Read prior audit results:
279
- - `{PROJECT_PATH}/critique/accessibility-audit.md`
280
- - `{PROJECT_PATH}/critique/accessibility-token-audit.md`
281
- - `{PROJECT_PATH}/review/accessibility-audit.md`
282
-
283
- If none exist, tell the user to run an audit first and stop.
284
-
285
- ### Generate statement
286
-
287
- Write `{PROJECT_PATH}/exports/accessibility-statement.md`:
288
-
289
- ```markdown
290
- # Accessibility Statement
291
-
292
- > Project: {name} | Generated: {DATE}
293
-
294
- ---
295
-
296
- ## Conformance Status
297
-
298
- **Target:** {level}
299
- **Status:** {Partially Conformant / Fully Conformant}
300
-
301
- This {project description} has been evaluated against {level} standards.
302
-
303
- ## Scope
304
-
305
- {Brief description of what was audited — design, code, or both}
306
-
307
- ## Known Limitations
308
-
309
- {List from audit findings — critical/major issues not yet resolved}
310
-
311
- - {Issue}: {brief description} — {planned resolution or workaround}
312
-
313
- ## Testing Methodology
314
-
315
- - Design audit: WCAG 2.2 checklist review of all screens
316
- - Token audit: Automated contrast ratio verification of all semantic color pairs
317
- - Code audit: Manual and grep-based review of ARIA, keyboard, semantic HTML
318
- - Tools used: {list from testing methodology}
319
-
320
- ## Feedback
321
-
322
- If you encounter accessibility barriers, please contact:
323
-
324
- - **Email:** [placeholder@example.com]
325
- - **Response time:** [X business days]
326
-
327
- ## Assessment Date
328
-
329
- Last reviewed: {DATE}
330
- ```
331
-
332
- ### Completion
333
-
334
- Display result and use `AskUserQuestion`:
335
- - **View statement** — "read the accessibility statement"
336
- - **Done** — "that's all for now"
337
-
338
- ## Step 8: Update STATE.md
207
+ ## Step 5: Update STATE.md
339
208
 
340
209
  If within a project and files were written:
341
210
  - Read `{PROJECT_PATH}/STATE.md`
342
211
  - Note accessibility audit completion in the relevant phase section
343
212
  - Do not change phase status — accessibility is a supplementary check
344
-
345
- ## Step 9: Completion output
346
-
347
- If not already displayed by a mode-specific step above, display the appropriate completion output and `AskUserQuestion` routing.
348
213
  </process>
@@ -0,0 +1,218 @@
1
+ ---
2
+ name: accessibility-audit
3
+ description: Full WCAG accessibility audit — design screens, codebase, or generate compliance statement
4
+ user-invocable: true
5
+ allowed-tools:
6
+ - Read
7
+ - Write
8
+ - Edit
9
+ - Bash
10
+ - Agent
11
+ - Glob
12
+ - Grep
13
+ - AskUserQuestion
14
+ ---
15
+ <context>
16
+ Full accessibility audit skill that spawns the `gsp-accessibility-auditor` agent for deep WCAG analysis. Works two ways:
17
+ 1. **Standalone** — user runs `/gsp:accessibility-audit` directly for design, code, or statement generation
18
+ 2. **As a building block** — critique and review phases detect prior accessibility output and reuse it
19
+
20
+ For quick contrast checks or token audits, use `/gsp:accessibility` instead.
21
+
22
+ Follows the composable pattern: deterministic modes, predictable output paths, filesystem as integration layer.
23
+ </context>
24
+
25
+ <objective>
26
+ Run full accessibility audits — design screen reviews, codebase ARIA/keyboard/semantic checks, or compliance statement generation.
27
+
28
+ **Input:** Mode flag + optional arguments
29
+ **Output:** Audit chunks and fix lists in the appropriate project directory
30
+ **Agent:** `gsp-accessibility-auditor` (for design and code modes), inline for statement
31
+ </objective>
32
+
33
+ <execution_context>
34
+ @${CLAUDE_SKILL_DIR}/../../prompts/08-accessibility-auditor.md
35
+ @${CLAUDE_SKILL_DIR}/../../references/wcag-checklist.md
36
+ </execution_context>
37
+
38
+ <rules>
39
+ - Always use `AskUserQuestion` for user interaction — never prompt via plain text
40
+ - Statement mode reads prior audit results — fails gracefully if none exist
41
+ - Default conformance level is AA unless overridden by `--level AAA` or config
42
+ - Foundation chunks follow `references/chunk-format.md` format
43
+ </rules>
44
+
45
+ <process>
46
+ ## Step 1: Parse invocation
47
+
48
+ Read `$ARGUMENTS` to determine the mode:
49
+
50
+ | Input | Mode | Agent? | Output |
51
+ |-------|------|--------|--------|
52
+ | (no args) | Design audit on `.design/` chunks | Yes (`gsp-accessibility-auditor`) | `critique/accessibility-audit.md` + `accessibility-fixes.md` |
53
+ | `--code` | Codebase audit: ARIA, keyboard, semantic HTML | Yes (`gsp-accessibility-auditor`) | `review/accessibility-audit.md` + `accessibility-fixes.md` |
54
+ | `--statement` | Generate accessibility statement from prior audits | No (inline) | `exports/accessibility-statement.md` |
55
+
56
+ Additional flag: `--level AAA` overrides conformance level (default: AA).
57
+
58
+ ## Step 2: Resolve context
59
+
60
+ Resolve project from `.design/projects/` (one → use it, multiple → ask). Set `PROJECT_PATH`.
61
+
62
+ Read `{PROJECT_PATH}/config.json` to get:
63
+ - `accessibility_level` — override conformance level (if not set via `--level` flag)
64
+ - `implementation_target` — needed for code mode
65
+
66
+ Read `{PROJECT_PATH}/brand.ref` to resolve brand path:
67
+ - Set `BRAND_PATH` = `.design/branding/{brand}`
68
+
69
+ Determine final conformance level:
70
+ 1. `--level` flag (highest priority)
71
+ 2. `accessibility_level` from config.json
72
+ 3. Default: "WCAG 2.2 AA"
73
+
74
+ ## Step 3: Design audit mode (default, no flags)
75
+
76
+ Verify design chunks exist:
77
+ - Read `{PROJECT_PATH}/design/INDEX.md` to find screen chunks
78
+ - If no design chunks, tell user to complete design phase first and stop
79
+
80
+ ### Spawn agent
81
+
82
+ Spawn `gsp-accessibility-auditor` with:
83
+ - All design chunks from `{PROJECT_PATH}/design/`
84
+ - Brand identity context (color system, typography)
85
+ - Brand system context (tokens, components)
86
+ - Conformance level
87
+ - WCAG checklist reference
88
+ - **Output path:** `{PROJECT_PATH}/critique/`
89
+ - **Instructions:** "Audit all design screens against {level}. Write `accessibility-audit.md` and `accessibility-fixes.md` to the output path."
90
+
91
+ ### Completion
92
+
93
+ Display result:
94
+
95
+ ```
96
+ /gsp:accessibility-audit — design audit complete
97
+ ═══════════════════════════════════════
98
+
99
+ {PROJECT_PATH}/critique/
100
+ ├── accessibility-audit.md
101
+ └── accessibility-fixes.md
102
+
103
+ ─────────────────────────────────────
104
+ ```
105
+
106
+ Use `AskUserQuestion`:
107
+ - **Run token audit** — "run `/gsp:accessibility --tokens` to check design token contrast pairs"
108
+ - **Continue to build** — "implement designs in the codebase"
109
+ - **View audit** — "read the accessibility report"
110
+ - **Done** — "that's all for now"
111
+
112
+ ## Step 4: Code audit mode (`--code`)
113
+
114
+ Determine codebase scope:
115
+ - Read `{PROJECT_PATH}/config.json` for `implementation_target`
116
+ - If build phase completed, read `{PROJECT_PATH}/build/BUILD-LOG.md` for file paths
117
+ - Otherwise, use `implementation_target` to determine where to look
118
+
119
+ ### Spawn agent
120
+
121
+ Spawn `gsp-accessibility-auditor` with:
122
+ - Codebase paths to audit
123
+ - Brand system tokens (for contrast verification against hardcoded values)
124
+ - Conformance level
125
+ - WCAG checklist reference
126
+ - **Output path:** `{PROJECT_PATH}/review/`
127
+ - **Instructions:** "Code audit mode. Use Grep and Glob to find accessibility issues in the codebase. Check ARIA, keyboard handlers, semantic HTML, heading hierarchy, alt text, lang attributes, skip-nav, focus management. Write `accessibility-audit.md` and `accessibility-fixes.md` to the output path with actual file paths and line numbers."
128
+
129
+ ### Completion
130
+
131
+ Display result:
132
+
133
+ ```
134
+ /gsp:accessibility-audit --code — code audit complete
135
+ ═══════════════════════════════════════
136
+
137
+ {PROJECT_PATH}/review/
138
+ ├── accessibility-audit.md
139
+ └── accessibility-fixes.md
140
+
141
+ ─────────────────────────────────────
142
+ ```
143
+
144
+ Use `AskUserQuestion`:
145
+ - **Fix issues** — "address the accessibility issues found"
146
+ - **Generate statement** — "create an accessibility statement"
147
+ - **View audit** — "read the code accessibility report"
148
+ - **Done** — "that's all for now"
149
+
150
+ ## Step 5: Statement mode (`--statement`)
151
+
152
+ Read prior audit results:
153
+ - `{PROJECT_PATH}/critique/accessibility-audit.md`
154
+ - `{PROJECT_PATH}/critique/accessibility-token-audit.md`
155
+ - `{PROJECT_PATH}/review/accessibility-audit.md`
156
+
157
+ If none exist, tell the user to run an audit first and stop.
158
+
159
+ ### Generate statement
160
+
161
+ Write `{PROJECT_PATH}/exports/accessibility-statement.md`:
162
+
163
+ ```markdown
164
+ # Accessibility Statement
165
+
166
+ > Project: {name} | Generated: {DATE}
167
+
168
+ ---
169
+
170
+ ## Conformance Status
171
+
172
+ **Target:** {level}
173
+ **Status:** {Partially Conformant / Fully Conformant}
174
+
175
+ This {project description} has been evaluated against {level} standards.
176
+
177
+ ## Scope
178
+
179
+ {Brief description of what was audited — design, code, or both}
180
+
181
+ ## Known Limitations
182
+
183
+ {List from audit findings — critical/major issues not yet resolved}
184
+
185
+ - {Issue}: {brief description} — {planned resolution or workaround}
186
+
187
+ ## Testing Methodology
188
+
189
+ - Design audit: WCAG 2.2 checklist review of all screens
190
+ - Token audit: Automated contrast ratio verification of all semantic color pairs
191
+ - Code audit: Manual and grep-based review of ARIA, keyboard, semantic HTML
192
+ - Tools used: {list from testing methodology}
193
+
194
+ ## Feedback
195
+
196
+ If you encounter accessibility barriers, please contact:
197
+
198
+ - **Email:** [placeholder@example.com]
199
+ - **Response time:** [X business days]
200
+
201
+ ## Assessment Date
202
+
203
+ Last reviewed: {DATE}
204
+ ```
205
+
206
+ ### Completion
207
+
208
+ Display result and use `AskUserQuestion`:
209
+ - **View statement** — "read the accessibility statement"
210
+ - **Done** — "that's all for now"
211
+
212
+ ## Step 6: Update STATE.md
213
+
214
+ If within a project and files were written:
215
+ - Read `{PROJECT_PATH}/STATE.md`
216
+ - Note accessibility audit completion in the relevant phase section
217
+ - Do not change phase status — accessibility is a supplementary check
218
+ </process>
@@ -36,9 +36,7 @@ Audit an existing brand. Produce evolution map that guides research, strategy, a
36
36
  <process>
37
37
  ## Step 1: Resolve brand
38
38
 
39
- Scan `.design/branding/` for brand directories. One brand → use it. Multipleuse `AskUserQuestion`.
40
-
41
- Set `BRAND_PATH` = `.design/branding/{brand}`
39
+ Resolve brand from `.design/branding/` (one → use it, multipleask). Set `BRAND_PATH`.
42
40
 
43
41
  Read `{BRAND_PATH}/BRIEF.md` for aspirational direction.
44
42
  Read `{BRAND_PATH}/config.json` to confirm `brand_mode` is `evolve`.
@@ -38,9 +38,7 @@ Build the brand's visual identity.
38
38
  <process>
39
39
  ## Step 0: Resolve brand
40
40
 
41
- Scan `.design/branding/` for brand directories. One brand → use it. Multipleuse `AskUserQuestion`.
42
-
43
- Set `BRAND_PATH` = `.design/branding/{brand}`
41
+ Resolve brand from `.design/branding/` (one → use it, multipleask). Set `BRAND_PATH`.
44
42
  If missing, tell user to run `/gsp:start` first.
45
43
 
46
44
  ## Step 1: Validate prerequisites
@@ -33,10 +33,7 @@ Build the design system, generate brand guidelines, and complete the branding di
33
33
  <process>
34
34
  ## Step 0: Resolve brand
35
35
 
36
- Scan `.design/branding/` for brand directories. If only one brand exists, use it. If multiple, ask the user which brand to work on.
37
-
38
- Set `BRAND_PATH` = `.design/branding/{brand}`
39
-
36
+ Resolve brand from `.design/branding/` (one use it, multiple ask). Set `BRAND_PATH`.
40
37
  If BRAND_PATH doesn't exist, tell the user to run `/gsp:start` first.
41
38
 
42
39
  ## Step 1: Load context
@@ -45,14 +42,11 @@ If BRAND_PATH doesn't exist, tell the user to run `/gsp:start` first.
45
42
 
46
43
  Read `{BRAND_PATH}/identity/INDEX.md`. If it exists, load all identity chunks + `palettes.json`.
47
44
 
48
- Fallback: read `{BRAND_PATH}/identity/IDENTITY.md` (legacy monolith) + `palettes.json`. Log: "⚠️ Legacy identity format detected consider re-running /gsp:brand-identity for chunk output."
49
-
50
- If neither exists, check if identity phase is complete in brand STATE.md. If not, tell the user to complete brand identity first (run `/gsp:brand-identity`).
45
+ If INDEX.md doesn't exist, check if identity phase is complete in brand STATE.md. If not, tell the user to complete brand identity first (run `/gsp:brand-identity`).
51
46
 
52
47
  ### Strategy (selective, chunk-first)
53
48
 
54
49
  Read `{BRAND_PATH}/strategy/INDEX.md`. If it exists, load selective chunks (voice-and-tone.md, brand-platform.md).
55
- Fallback: read `{BRAND_PATH}/strategy/STRATEGY.md`.
56
50
 
57
51
  ### Brand context
58
52
 
@@ -109,7 +103,7 @@ Redesign the system from the ground up, informed by what exists.
109
103
  ## Step 3: Spawn pattern architect — Pass 1: Foundations
110
104
 
111
105
  Spawn the `gsp-pattern-architect` agent with:
112
- - All identity chunks (or IDENTITY.md fallback) + palettes.json
106
+ - All identity chunks + palettes.json
113
107
  - The BRIEF.md content
114
108
  - The Design System Architect prompt (01)
115
109
  - The patterns output template