@se-studio/project-build 1.0.105 → 1.0.106

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.
package/CHANGELOG.md CHANGED
@@ -1,5 +1,11 @@
1
1
  # @se-studio/project-build
2
2
 
3
+ ## 1.0.106
4
+
5
+ ### Patch Changes
6
+
7
+ - Bulk version bump: patch for all packages
8
+
3
9
  ## 1.0.105
4
10
 
5
11
  ### Patch Changes
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@se-studio/project-build",
3
- "version": "1.0.105",
3
+ "version": "1.0.106",
4
4
  "description": "Build tools and management scripts for SE Studio projects",
5
5
  "repository": {
6
6
  "type": "git",
@@ -0,0 +1,282 @@
1
+ ---
2
+ name: generate-all-guidelines
3
+ description: Generates CMS guidelines for every component, collection, and external in one automated run. Handles bulk screenshot capture upfront, then uses parallel subagents for guideline generation. Use when setting up a new project or after significant changes across many component types.
4
+ license: Private
5
+ metadata:
6
+ author: se-core-product
7
+ version: "1.0.0"
8
+ ---
9
+
10
+ # Generate All Guidelines (Full Site Run)
11
+
12
+ Generates CMS guidelines for every registered type in one orchestrated run.
13
+ The workflow front-loads all screenshot capture so parallel subagents can go
14
+ straight to guideline writing without waiting for screenshots one by one.
15
+
16
+ ## When to use
17
+
18
+ - First-time guideline generation for a new project
19
+ - Bulk regeneration after major refactoring
20
+ - Adding screenshot links to an existing set of guidelines that has no screenshots yet
21
+
22
+ Use the **generate-cms-guidelines** skill (`single` or `merge-only` mode) for incremental updates after this initial run.
23
+
24
+ ---
25
+
26
+ ## Prerequisites
27
+
28
+ All of these must be true before starting. Check each one.
29
+
30
+ 1. **Dev server is running** at the app's port. Confirm:
31
+ ```bash
32
+ curl -sL http://localhost:<PORT>/api/cms/discovery/ | head -c 100
33
+ ```
34
+ Should return JSON starting with `{"components":[`.
35
+
36
+ 2. **`agent-browser` is installed** (needed for screenshot capture):
37
+ ```bash
38
+ agent-browser --version
39
+ # If missing: npm install -g agent-browser && agent-browser install
40
+ ```
41
+
42
+ 3. **Showcase pipeline completed** — `src/generated/showcase-mocks.json` must contain real content (not just `{}` stubs). If empty, run the **curate-showcase-mocks** skill first.
43
+
44
+ 4. **Field list generated:**
45
+ ```bash
46
+ pnpm run cms-discovery:field-list
47
+ # or from repo root: pnpm --filter <appName> exec cms-generate-field-list --app-dir .
48
+ ```
49
+ Produces `src/generated/cms-discovery/field-list.json`.
50
+
51
+ 5. **LLM pre-pass run** — `src/generated/cms-discovery/accepted-variants/` must contain JSON files (one per component type). These are written by `generate-showcase-mocks`. If the directory is empty or missing, run:
52
+ ```bash
53
+ pnpm generate-showcase-mocks
54
+ # or: node node_modules/@se-studio/project-build/dist/generate-showcase-mocks.js
55
+ ```
56
+
57
+ ---
58
+
59
+ ## Inputs
60
+
61
+ | Input | Example |
62
+ |-------|---------|
63
+ | App directory (absolute or repo-relative) | `apps/example-se2026` or `/home/nick/source/se/se-website-2026` |
64
+ | Discovery URL | `http://localhost:3012/api/cms/discovery/` |
65
+ | Port | `3012` |
66
+
67
+ ---
68
+
69
+ ## Pipeline Overview
70
+
71
+ ```
72
+ Phase 0 — Discover Fetch all type names from the discovery API
73
+ Phase 1 — Screenshots Bulk capture all component screenshots in one loop (CLI only, no LLM)
74
+ Phase 2 — Collections Generate collection + external guideline fragments (CLI only)
75
+ Phase 3 — Components Parallel subagents: Phase B (write) + C (colour) + D (validate) per type
76
+ Phase 4 — Merge cms-merge-guidelines (single CLI call)
77
+ Phase 5 — Commit git add + commit
78
+ ```
79
+
80
+ Phases 1 and 2 are purely mechanical — they run fast and require no LLM supervision.
81
+ Phase 3 is where parallel subagents do the guideline writing.
82
+
83
+ ---
84
+
85
+ ## Phase 0 — Discover all types
86
+
87
+ Fetch the discovery endpoint and collect the full list of component type names.
88
+ Store this as your working list.
89
+
90
+ ```bash
91
+ curl -sL http://localhost:<PORT>/api/cms/discovery/ > /tmp/discovery.json
92
+ ```
93
+
94
+ From the JSON:
95
+ - `data.components[].name` — component types needing Phases 1 + 3
96
+ - `data.collections[].name` — collection types (Phase 2 handles these)
97
+ - `data.externals[].name` — external types (Phase 2 handles these)
98
+
99
+ Compute each component's slug: `typeName.toLowerCase().replace(/\s+/g, '-').replace(/[()]/g, '')`
100
+
101
+ **Skip check:** For each component, check if BOTH of these already exist:
102
+ - `<appDir>/docs/cms-guidelines/components/<slug>.md`
103
+ - `<appDir>/docs/cms-guidelines/screenshots/index.json` has an entry for this type
104
+
105
+ If both exist, skip that component (already done). Only process the unfinished ones.
106
+ To force-regenerate all, ignore this check.
107
+
108
+ ---
109
+
110
+ ## Phase 1 — Bulk screenshot capture (all components at once)
111
+
112
+ This is the key step that makes the full run fast. Instead of capturing screenshots one
113
+ type at a time inside each subagent, we capture everything upfront in a tight loop.
114
+
115
+ For each component type name that needs processing, run:
116
+
117
+ ```bash
118
+ # Check if accepted-variants file exists for this type
119
+ SLUG="<type-slug>"
120
+ VARIANTS_FILE="<appDir>/src/generated/cms-discovery/accepted-variants/${SLUG}.json"
121
+
122
+ if [ -f "$VARIANTS_FILE" ]; then
123
+ cms-capture-screenshots --variants "$VARIANTS_FILE" --app-dir <appDir>
124
+ else
125
+ # Fall back to default variants for the type
126
+ cms-capture-screenshots --type "<TYPE_NAME>" --app-dir <appDir>
127
+ fi
128
+ ```
129
+
130
+ Run all of these sequentially (screenshots require the browser to be driven one at a time).
131
+ `cms-capture-screenshots` automatically updates `docs/cms-guidelines/screenshots/index.json` after each run.
132
+
133
+ **If `agent-browser` is not available:** the command degrades to printing URLs only.
134
+ Install it first: `npm install -g agent-browser && agent-browser install`.
135
+
136
+ After all captures, confirm `docs/cms-guidelines/screenshots/index.json` exists and has entries.
137
+
138
+ ---
139
+
140
+ ## Phase 2 — Collections and externals (CLI, no subagents needed)
141
+
142
+ Run once — this regenerates all collection guidelines and creates any missing external templates:
143
+
144
+ ```bash
145
+ cms-generate-collection-guidelines \
146
+ --discovery-url http://localhost:<PORT>/api/cms/discovery/ \
147
+ --app-dir <appDir>
148
+ ```
149
+
150
+ This writes `docs/cms-guidelines/collections/*.md` and creates stub `docs/cms-guidelines/externals/*.md` for any new externals.
151
+
152
+ **External schemas:** If any new `externals/*.md` stubs were created, fill in their
153
+ `## Fields & Schema` and `## Usage` sections manually before continuing (or after the full run).
154
+ The rest of the pipeline doesn't depend on them being complete.
155
+
156
+ ---
157
+
158
+ ## Phase 3 — Component guidelines (parallel subagents)
159
+
160
+ Launch subagents in batches of **4 in parallel**. Each subagent handles one component type
161
+ through Phases B, C, and D only (screenshots are already done from Phase 1).
162
+
163
+ ### Subagent prompt template
164
+
165
+ For each component type, use this prompt (fill in `<COMPONENT_TYPE>`, `<APP_DIR>`, `<PORT>`):
166
+
167
+ ```
168
+ Generate the CMS guideline fragment for <COMPONENT_TYPE>.
169
+ Target app directory: <APP_DIR>
170
+
171
+ IMPORTANT: Screenshots have already been captured. Skip Phase A entirely.
172
+ Go straight to Phase B.
173
+
174
+ Files to read before starting:
175
+ - packages/contentful-cms/skills/cms-guidelines/generation-prompt.md (Phase B instructions)
176
+ - packages/contentful-cms/skills/cms-guidelines/colour-hint-prompt.md (Phase C instructions)
177
+ - packages/contentful-cms/skills/cms-guidelines/validation-prompt.md (Phase D instructions)
178
+ - <APP_DIR>/src/project/components/<TypeName>.tsx (component source)
179
+ - <APP_DIR>/src/generated/cms-discovery/field-list.json
180
+ - <APP_DIR>/src/generated/cms-discovery/accepted-variants/<type-slug>.json
181
+ - <APP_DIR>/docs/cms-guidelines/screenshots/index.json (which screenshots exist)
182
+ - <APP_DIR>/.contentful-cms.json (get devBaseUrl for screenshot links)
183
+
184
+ === Phase B: Generate guideline ===
185
+ Follow generation-prompt.md (three-step process).
186
+ The guideline MUST include a ## Screenshots table using devBaseUrl from .contentful-cms.json.
187
+ Only include variants present in screenshots/index.json.
188
+ Save output to <APP_DIR>/docs/cms-guidelines/components/<type-slug>.md.
189
+
190
+ === Phase C: Colour hint ===
191
+ Follow colour-hint-prompt.md.
192
+ Run:
193
+ cms-update-colour-hints --app-dir <APP_DIR> \
194
+ --type "<COMPONENT_TYPE>" --hint "<hint>" --notes "<brief reason>"
195
+
196
+ === Phase D: Validate ===
197
+ Follow validation-prompt.md. Apply fixes until APPROVED.
198
+
199
+ Do NOT run merge — that runs once after all components complete.
200
+ ```
201
+
202
+ ### Batch execution
203
+
204
+ Process all unfinished components in batches:
205
+ - Launch 4 subagents simultaneously
206
+ - Wait for all 4 to complete before launching the next batch
207
+ - If any subagent fails or times out, note the type name and re-run it individually
208
+ using the **generate-cms-guidelines** skill in `single` mode
209
+
210
+ Typical time per component: 3–6 minutes (mostly LLM time for generation + validation).
211
+ Typical total time for 15 components in batches of 4: ~25–35 minutes.
212
+
213
+ ---
214
+
215
+ ## Phase 4 — Merge
216
+
217
+ After all subagents complete, rebuild the combined document:
218
+
219
+ ```bash
220
+ cms-merge-guidelines --app-dir <appDir>
221
+ # or from the app directory:
222
+ pnpm run cms-guidelines:merge
223
+ ```
224
+
225
+ Open `docs/cms-guidelines/COMPONENT_GUIDELINES_FOR_LLM.md` and do a quick sanity check:
226
+ - Does it contain all expected types?
227
+ - Do the screenshot image links look correct (using the right port)?
228
+
229
+ ---
230
+
231
+ ## Phase 5 — Commit
232
+
233
+ ```bash
234
+ git add <appDir>/docs/cms-guidelines/ <appDir>/src/generated/cms-discovery/
235
+ git commit -m "chore(<appName>): generate all CMS guidelines with screenshots"
236
+ ```
237
+
238
+ ---
239
+
240
+ ## Handling failures and partial runs
241
+
242
+ The pipeline is designed to be resumable. If it stops partway through:
243
+
244
+ - **Phase 1 failed for some types:** Re-run `cms-capture-screenshots` for just those types.
245
+ `index.json` is additive — it won't remove entries from previous runs.
246
+ - **Phase 3 subagent failed:** Use `generate-cms-guidelines` skill in `single` mode for that type.
247
+ - **Phase 2 not run yet:** Re-run `cms-generate-collection-guidelines` — it's idempotent for collections.
248
+ - **Phase 4 merge not run:** Just run `cms-merge-guidelines` — it reads whatever `.md` files exist.
249
+
250
+ To re-run everything for a single type after fixing a problem:
251
+ ```bash
252
+ # Re-capture screenshots
253
+ cms-capture-screenshots --type "<TYPE_NAME>" --app-dir <appDir>
254
+ # Then use generate-cms-guidelines skill (single mode) for the guideline
255
+ ```
256
+
257
+ ---
258
+
259
+ ## Quick reference — full command sequence
260
+
261
+ ```bash
262
+ # 0. Prerequisites check
263
+ curl -sL http://localhost:<PORT>/api/cms/discovery/ | head -c 100
264
+ agent-browser --version
265
+
266
+ # 1. Bulk screenshots (repeat for each component type)
267
+ cms-capture-screenshots --variants src/generated/cms-discovery/accepted-variants/<slug>.json --app-dir <appDir>
268
+
269
+ # 2. Collections + externals
270
+ cms-generate-collection-guidelines \
271
+ --discovery-url http://localhost:<PORT>/api/cms/discovery/ \
272
+ --app-dir <appDir>
273
+
274
+ # 3. [Parallel subagents — see Phase 3 above]
275
+
276
+ # 4. Merge
277
+ cms-merge-guidelines --app-dir <appDir>
278
+
279
+ # 5. Commit
280
+ git add docs/cms-guidelines/ src/generated/cms-discovery/
281
+ git commit -m "chore: generate all CMS guidelines"
282
+ ```
@@ -22,7 +22,7 @@ Guidelines are only meaningful when the showcase is populated with realistic dat
22
22
  ```bash
23
23
  pnpm --filter <appName> generate-showcase
24
24
  ```
25
- This writes `src/cms-showcase/showcase-examples.json` to the app.
25
+ This writes `src/generated/showcase-examples.json` to the app.
26
26
  3. **Curate showcase mocks:** Follow the [curate-showcase-mocks skill](.cursor/skills/se-marketing-sites/curate-showcase-mocks/SKILL.md) to produce `showcase-mocks.json`.
27
27
  4. **Generate field list:**
28
28
  ```bash
@@ -89,7 +89,7 @@ export type CmsColourName = NonNullable<TypeComponentFields['backgroundColour']>
89
89
 
90
90
  ## The Showcase
91
91
 
92
- Registering a component automatically adds it to the CMS Showcase (usually available at `/cms-showcase` in development). This allows you to view the component in isolation using its mock data.
92
+ Registering a component automatically adds it to the CMS Showcase (usually available at `/cms/showcase` in development). This allows you to view the component in isolation using its mock data.
93
93
 
94
94
  * Ensure your `mock` object in the registration is complete and realistic.
95
95
  * The showcase helps verify that `usedFields` and `UnusedChecker` are working correctly.