start-vibing 4.4.12 → 4.4.13
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/package.json +1 -1
- package/template/.claude/CLAUDE.md +117 -92
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "start-vibing",
|
|
3
|
-
"version": "4.4.
|
|
3
|
+
"version": "4.4.13",
|
|
4
4
|
"description": "Setup Claude Code with 9 plugins, 6 community skills, and 8 MCP servers. Parallel install, auto-accept, superpowers + ralph-loop. e2e-audit 0.2.0 refactor (skill-only, no agents): SessionStart hook + slash command make the skill keyword-invokable (\"e2e audit\", \"roda o e2e\", \"integration test\", \"test coverage gaps\"). Source-first discovery via detect-stack, discover-routes (Next app/pages/Remix/SvelteKit/Nuxt/Astro), discover-api-surface (HTTP handlers, tRPC procedures, GraphQL, server actions, middleware auth), inventory-existing-tests (preserve prior corpus + sha256 drift hash), and detect-uncovered (branch-diff vs origin/main finds changes not covered by existing specs). Report-then-ask between mapping and Playwright run; post-run-feedback report before writing findings. SHOT+TRACE+ASSERT+SOURCE evidence quad per non-meta finding; meta rules (coverage-gap-*, uncovered-*, test-drift, stack-detect, post-run-feedback) exempt. verify-audit.sh enforces schema + quad. Generic (no project leakage). super-design 0.7.0 carries over.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"bin": {
|
|
@@ -5,16 +5,34 @@ For user-facing project rules, see `/CLAUDE.md`.
|
|
|
5
5
|
|
|
6
6
|
---
|
|
7
7
|
|
|
8
|
+
## Language Policy (HARD RULE)
|
|
9
|
+
|
|
10
|
+
> **ENGLISH ONLY — for everything Claude produces — UNLESS the user explicitly asks for another language in the current request.**
|
|
11
|
+
|
|
12
|
+
Applies to **every** output, no exceptions beyond an explicit user override:
|
|
13
|
+
|
|
14
|
+
- Chat replies, status updates, summaries, questions back to the user
|
|
15
|
+
- Code (identifiers, variables, function/class names, type names)
|
|
16
|
+
- Comments, JSDoc, docstrings
|
|
17
|
+
- Commit messages, PR titles, PR descriptions, branch names
|
|
18
|
+
- Documentation (`/docs/**`, README, CLAUDE.md updates, changelogs, ADRs)
|
|
19
|
+
- Test names, error messages, log messages, CLI help strings
|
|
20
|
+
- File and folder names, configuration keys
|
|
21
|
+
|
|
22
|
+
The user may write to Claude in Portuguese, Spanish, or any other language — Claude still responds and writes in English unless the user explicitly says e.g. "responda em português", "write this in Spanish". An override applies only to the current request; revert to English on the next turn. Never mix English + another language in the same output. If unsure whether a request is an override, default to English and ask one short clarifying question.
|
|
23
|
+
|
|
24
|
+
---
|
|
25
|
+
|
|
8
26
|
## What start-vibing v4 Installs
|
|
9
27
|
|
|
10
28
|
start-vibing is a CLI (`npx start-vibing`) that sets up Claude Code with a complete development system in ~30 seconds:
|
|
11
29
|
|
|
12
|
-
| Component
|
|
13
|
-
|
|
14
|
-
| **MCP Servers**
|
|
15
|
-
| **Plugins**
|
|
16
|
-
| **Community Skills** | 5
|
|
17
|
-
| **Template Files**
|
|
30
|
+
| Component | Count | Installation Method |
|
|
31
|
+
| -------------------- | ----- | ---------------------------------------------------- |
|
|
32
|
+
| **MCP Servers** | 8 | `claude mcp add` (parallel, ~20s) |
|
|
33
|
+
| **Plugins** | 9 | `claude plugin install` (parallel, auto-accept, ~3s) |
|
|
34
|
+
| **Community Skills** | 5 | Direct HTTPS download from GitHub (~0.3s) |
|
|
35
|
+
| **Template Files** | ~30 | File copy (agents, skills, config, CLAUDE.md) |
|
|
18
36
|
|
|
19
37
|
### Installation Architecture
|
|
20
38
|
|
|
@@ -32,14 +50,14 @@ All phases run best-effort. MCP and plugin failures are non-blocking — `settin
|
|
|
32
50
|
|
|
33
51
|
### CLI Options
|
|
34
52
|
|
|
35
|
-
| Flag
|
|
36
|
-
|
|
37
|
-
| `--force`
|
|
38
|
-
| `--new`
|
|
39
|
-
| `--no-claude`
|
|
40
|
-
| `--no-mcp`
|
|
41
|
-
| `--no-skills`
|
|
42
|
-
| `--no-update-check` | Skip version check
|
|
53
|
+
| Flag | Effect |
|
|
54
|
+
| ------------------- | ------------------------------------------------- |
|
|
55
|
+
| `--force` | Overwrite all files (including custom ones) |
|
|
56
|
+
| `--new` | Start fresh Claude session (default: resume last) |
|
|
57
|
+
| `--no-claude` | Skip Claude Code installation and launch |
|
|
58
|
+
| `--no-mcp` | Skip MCP server installation |
|
|
59
|
+
| `--no-skills` | Skip community skills installation |
|
|
60
|
+
| `--no-update-check` | Skip version check |
|
|
43
61
|
|
|
44
62
|
---
|
|
45
63
|
|
|
@@ -58,13 +76,13 @@ All phases run best-effort. MCP and plugin failures are non-blocking — `settin
|
|
|
58
76
|
|
|
59
77
|
## Agents (5 Active Subagents)
|
|
60
78
|
|
|
61
|
-
| Agent
|
|
62
|
-
|
|
63
|
-
| **research-web**
|
|
64
|
-
| **documenter**
|
|
65
|
-
| **commit-manager**
|
|
66
|
-
| **claude-md-compactor** | sonnet | Compacts CLAUDE.md when it exceeds 40k chars
|
|
67
|
-
| **tester-unit**
|
|
79
|
+
| Agent | Model | Purpose |
|
|
80
|
+
| ----------------------- | ------ | -------------------------------------------------------------------------------------------------- |
|
|
81
|
+
| **research-web** | sonnet | Researches best practices (2025-2026) with ontology-based structuring, output to `/docs/research/` |
|
|
82
|
+
| **documenter** | sonnet | Analyzes sessions via git log/diff, writes changelog + technical docs + ADRs to `/docs/` |
|
|
83
|
+
| **commit-manager** | haiku | Manages git commits, conventional format, merge workflow |
|
|
84
|
+
| **claude-md-compactor** | sonnet | Compacts CLAUDE.md when it exceeds 40k chars |
|
|
85
|
+
| **tester-unit** | sonnet | Creates unit tests with Vitest for new functions and utilities |
|
|
68
86
|
|
|
69
87
|
### Agent Workflow Order
|
|
70
88
|
|
|
@@ -79,6 +97,7 @@ Agents are dispatched via the `Agent` tool with `subagent_type` matching agent n
|
|
|
79
97
|
The research-web agent outputs findings to `/docs/research/[topic].md` (NOT to `.claude/skills/research-cache/`).
|
|
80
98
|
|
|
81
99
|
**Research flow:**
|
|
100
|
+
|
|
82
101
|
1. Check `/docs/research/` for existing findings (reuse if fresh < 3 months)
|
|
83
102
|
2. Build ontology map (concepts → relationships → constraints)
|
|
84
103
|
3. Search with `[topic] [aspect] [2025-2026] [context]` queries
|
|
@@ -86,6 +105,7 @@ The research-web agent outputs findings to `/docs/research/[topic].md` (NOT to `
|
|
|
86
105
|
5. Save structured output to `/docs/research/`
|
|
87
106
|
|
|
88
107
|
**For UI/UX tasks, the agent also runs:**
|
|
108
|
+
|
|
89
109
|
- Competitor analysis (3-5 competitors, heuristic evaluation)
|
|
90
110
|
- Design system pattern check (shadcn/ui, Radix, WCAG 2.1)
|
|
91
111
|
- User flow mapping (happy path + 2 error paths)
|
|
@@ -96,6 +116,7 @@ The research-web agent outputs findings to `/docs/research/[topic].md` (NOT to `
|
|
|
96
116
|
The documenter agent runs after implementation, analyzes the session, and writes structured docs to `/docs/`.
|
|
97
117
|
|
|
98
118
|
**Documentation flow:**
|
|
119
|
+
|
|
99
120
|
1. Run `git log` + `git diff` to analyze what changed
|
|
100
121
|
2. Classify changes: per-commit, per-feature, per-session
|
|
101
122
|
3. Mini-research for technologies used (check `/docs/research/` first, else 1-2 queries)
|
|
@@ -103,6 +124,7 @@ The documenter agent runs after implementation, analyzes the session, and writes
|
|
|
103
124
|
5. Update all indexes (`/docs/index.md` + per-folder indexes)
|
|
104
125
|
|
|
105
126
|
**Output structure:**
|
|
127
|
+
|
|
106
128
|
```
|
|
107
129
|
/docs/
|
|
108
130
|
├── index.md # Root index
|
|
@@ -121,6 +143,7 @@ The documenter agent runs after implementation, analyzes the session, and writes
|
|
|
121
143
|
```
|
|
122
144
|
|
|
123
145
|
**Writing rules:**
|
|
146
|
+
|
|
124
147
|
- Self-contained sections (AI RAG chunk retrieval)
|
|
125
148
|
- What→Why→How progression (humans first)
|
|
126
149
|
- Before→After pattern for all changes (mandatory)
|
|
@@ -136,22 +159,22 @@ Plugins are the primary extension mechanism. All 9 are installed in parallel wit
|
|
|
136
159
|
|
|
137
160
|
### Core Workflow Plugins
|
|
138
161
|
|
|
139
|
-
| Plugin
|
|
140
|
-
|
|
141
|
-
| **superpowers**
|
|
142
|
-
| **ralph-loop**
|
|
143
|
-
| **context7**
|
|
144
|
-
| **code-simplifier** | Skill + command
|
|
162
|
+
| Plugin | Mechanism | Purpose |
|
|
163
|
+
| ------------------- | ------------------- | ------------------------------------------------------------------- |
|
|
164
|
+
| **superpowers** | Skills + commands | TDD, brainstorming, debugging, planning, code review, git worktrees |
|
|
165
|
+
| **ralph-loop** | Stop hook + command | Iterative autonomous development loop with checkpoints |
|
|
166
|
+
| **context7** | Skill + agent + MCP | Auto library docs — replaces manual context7 MCP server |
|
|
167
|
+
| **code-simplifier** | Skill + command | Refine code for clarity, reduce nesting, improve naming |
|
|
145
168
|
|
|
146
169
|
### Development Plugins
|
|
147
170
|
|
|
148
|
-
| Plugin
|
|
149
|
-
|
|
150
|
-
| **typescript-lsp**
|
|
151
|
-
| **security-guidance** | PreToolUse hook (auto)
|
|
152
|
-
| **code-review**
|
|
153
|
-
| **commit-commands**
|
|
154
|
-
| **frontend-design**
|
|
171
|
+
| Plugin | Mechanism | Purpose |
|
|
172
|
+
| --------------------- | -------------------------- | -------------------------------------------------------- |
|
|
173
|
+
| **typescript-lsp** | LSP server (auto) | Type diagnostics, go-to-definition, hover info |
|
|
174
|
+
| **security-guidance** | PreToolUse hook (auto) | OWASP Top 10, vulnerability scan, blocks unsafe patterns |
|
|
175
|
+
| **code-review** | `/code-review` command | Code quality analysis, PR review |
|
|
176
|
+
| **commit-commands** | `/commit` command | Git commit, push, PR flows with conventional format |
|
|
177
|
+
| **frontend-design** | `/frontend-design` command | Production-grade UI design with competitor research |
|
|
155
178
|
|
|
156
179
|
### Plugin Installation Details
|
|
157
180
|
|
|
@@ -193,6 +216,7 @@ All file changes and git history persist between iterations. Best for multi-file
|
|
|
193
216
|
### Using Code Simplifier (POST-IMPLEMENTATION)
|
|
194
217
|
|
|
195
218
|
After implementing, run `/simplify` to:
|
|
219
|
+
|
|
196
220
|
- Reduce nesting and redundancy
|
|
197
221
|
- Improve naming and readability
|
|
198
222
|
- Replace nested ternaries with early returns
|
|
@@ -206,10 +230,10 @@ All 8 MCPs are installed in parallel via `claude mcp add -s user` (~20s total).
|
|
|
206
230
|
|
|
207
231
|
### Required MCPs (ALWAYS use these)
|
|
208
232
|
|
|
209
|
-
| Server
|
|
210
|
-
|
|
233
|
+
| Server | Package | Purpose |
|
|
234
|
+
| --------------------- | -------------------------------------------------- | ---------------------------------------------------------------------------- |
|
|
211
235
|
| `sequential-thinking` | `@modelcontextprotocol/server-sequential-thinking` | Structured reasoning for multi-step tasks, architecture decisions, debugging |
|
|
212
|
-
| `playwright`
|
|
236
|
+
| `playwright` | `@playwright/mcp@latest` | Browser automation for UI verification, E2E tests, visual validation |
|
|
213
237
|
|
|
214
238
|
> These 2 are **non-negotiable**. Skipping them leads to poor planning and untested UIs.
|
|
215
239
|
|
|
@@ -217,27 +241,27 @@ All 8 MCPs are installed in parallel via `claude mcp add -s user` (~20s total).
|
|
|
217
241
|
|
|
218
242
|
### Standard MCPs
|
|
219
243
|
|
|
220
|
-
| Server
|
|
221
|
-
|
|
222
|
-
| `memory`
|
|
223
|
-
| `nextjs-devtools` | `next-devtools-mcp@latest`
|
|
224
|
-
| `mongodb`
|
|
225
|
-
| `jira`
|
|
226
|
-
| `git`
|
|
227
|
-
| `fetch`
|
|
244
|
+
| Server | Package | Transport | Purpose |
|
|
245
|
+
| ----------------- | ------------------------------------- | ----------- | ------------------------------------------------ |
|
|
246
|
+
| `memory` | `@modelcontextprotocol/server-memory` | stdio (npx) | Persistent knowledge graph across sessions |
|
|
247
|
+
| `nextjs-devtools` | `next-devtools-mcp@latest` | stdio (npx) | Next.js runtime errors, routes, cache inspection |
|
|
248
|
+
| `mongodb` | `@mongodb-js/mongodb-mcp-server` | stdio (npx) | MongoDB queries, schema inspection, aggregation |
|
|
249
|
+
| `jira` | `@aashari/mcp-server-atlassian-jira` | stdio (npx) | Issue tracking, task management |
|
|
250
|
+
| `git` | `mcp-server-git` | stdio (uvx) | Git operations, repo search, history, diffs |
|
|
251
|
+
| `fetch` | `mcp-server-fetch` | stdio (uvx) | Fetch web pages as markdown |
|
|
228
252
|
|
|
229
253
|
### Optional MCPs (install manually)
|
|
230
254
|
|
|
231
255
|
These are shown to the user after setup but not auto-installed:
|
|
232
256
|
|
|
233
|
-
| Server
|
|
234
|
-
|
|
257
|
+
| Server | Install Command |
|
|
258
|
+
| -------- | ----------------------------------------------------------------------------------- |
|
|
235
259
|
| `github` | `claude mcp add --transport http -s user github https://api.githubcopilot.com/mcp/` |
|
|
236
|
-
| `sentry` | `claude mcp add --transport http -s user sentry https://mcp.sentry.dev/mcp`
|
|
237
|
-
| `figma`
|
|
238
|
-
| `linear` | `claude mcp add --transport http -s user linear https://mcp.linear.app/sse`
|
|
239
|
-
| `stripe` | `claude mcp add --transport http -s user stripe https://mcp.stripe.com`
|
|
240
|
-
| `vercel` | `claude mcp add --transport http -s user vercel https://mcp.vercel.com`
|
|
260
|
+
| `sentry` | `claude mcp add --transport http -s user sentry https://mcp.sentry.dev/mcp` |
|
|
261
|
+
| `figma` | `claude mcp add --transport http -s user figma https://mcp.figma.com/mcp` |
|
|
262
|
+
| `linear` | `claude mcp add --transport http -s user linear https://mcp.linear.app/sse` |
|
|
263
|
+
| `stripe` | `claude mcp add --transport http -s user stripe https://mcp.stripe.com` |
|
|
264
|
+
| `vercel` | `claude mcp add --transport http -s user vercel https://mcp.vercel.com` |
|
|
241
265
|
|
|
242
266
|
---
|
|
243
267
|
|
|
@@ -260,13 +284,13 @@ Community skills are downloaded directly from GitHub raw URLs (the `skillsadd` n
|
|
|
260
284
|
|
|
261
285
|
### Installed Skills
|
|
262
286
|
|
|
263
|
-
| Skill
|
|
264
|
-
|
|
265
|
-
| **react-best-practices**
|
|
266
|
-
| **web-design-guidelines** | `vercel-labs/agent-skills` | 100+ WCAG accessibility + UX audit rules
|
|
267
|
-
| **composition-patterns**
|
|
268
|
-
| **webapp-testing**
|
|
269
|
-
| **mcp-builder**
|
|
287
|
+
| Skill | Source Repo | Purpose |
|
|
288
|
+
| ------------------------- | -------------------------- | -------------------------------------------- |
|
|
289
|
+
| **react-best-practices** | `vercel-labs/agent-skills` | React/Next.js performance optimization rules |
|
|
290
|
+
| **web-design-guidelines** | `vercel-labs/agent-skills` | 100+ WCAG accessibility + UX audit rules |
|
|
291
|
+
| **composition-patterns** | `vercel-labs/agent-skills` | Compound component and composition patterns |
|
|
292
|
+
| **webapp-testing** | `anthropics/skills` | Real browser test execution with Playwright |
|
|
293
|
+
| **mcp-builder** | `anthropics/skills` | Guide for building MCP servers |
|
|
270
294
|
|
|
271
295
|
### Adding More Skills
|
|
272
296
|
|
|
@@ -294,12 +318,12 @@ curl -o .claude/skills/<name>/SKILL.md \
|
|
|
294
318
|
|
|
295
319
|
Project-specific settings in `.claude/config/`:
|
|
296
320
|
|
|
297
|
-
| File
|
|
298
|
-
|
|
299
|
-
| `project-config.json` | Stack, structure, commands
|
|
300
|
-
| `quality-gates.json`
|
|
321
|
+
| File | Purpose |
|
|
322
|
+
| --------------------- | ------------------------------ |
|
|
323
|
+
| `project-config.json` | Stack, structure, commands |
|
|
324
|
+
| `quality-gates.json` | Quality check commands |
|
|
301
325
|
| `testing-config.json` | Test framework and conventions |
|
|
302
|
-
| `security-rules.json` | Security audit rules
|
|
326
|
+
| `security-rules.json` | Security audit rules |
|
|
303
327
|
|
|
304
328
|
Agents read config files before acting. Do NOT hardcode project specifics — update the JSON files instead.
|
|
305
329
|
|
|
@@ -311,17 +335,17 @@ The `.claude/settings.json` file contains `enabledPlugins` which is the fallback
|
|
|
311
335
|
|
|
312
336
|
```json
|
|
313
337
|
{
|
|
314
|
-
|
|
315
|
-
|
|
316
|
-
|
|
317
|
-
|
|
318
|
-
|
|
319
|
-
|
|
320
|
-
|
|
321
|
-
|
|
322
|
-
|
|
323
|
-
|
|
324
|
-
|
|
338
|
+
"enabledPlugins": {
|
|
339
|
+
"typescript-lsp@claude-plugins-official": true,
|
|
340
|
+
"security-guidance@claude-plugins-official": true,
|
|
341
|
+
"code-review@claude-plugins-official": true,
|
|
342
|
+
"commit-commands@claude-plugins-official": true,
|
|
343
|
+
"frontend-design@claude-plugins-official": true,
|
|
344
|
+
"superpowers@claude-plugins-official": true,
|
|
345
|
+
"ralph-loop@claude-plugins-official": true,
|
|
346
|
+
"context7@claude-plugins-official": true,
|
|
347
|
+
"code-simplifier@claude-plugins-official": true
|
|
348
|
+
}
|
|
325
349
|
}
|
|
326
350
|
```
|
|
327
351
|
|
|
@@ -359,6 +383,7 @@ The `.claude/settings.json` file contains `enabledPlugins` which is the fallback
|
|
|
359
383
|
### After Completing Work
|
|
360
384
|
|
|
361
385
|
The documenter agent runs automatically after implementation:
|
|
386
|
+
|
|
362
387
|
1. Analyzes git log/diff for the session
|
|
363
388
|
2. Writes changelog + technical docs + ADRs as needed
|
|
364
389
|
3. Updates all indexes
|
|
@@ -397,23 +422,23 @@ All implementations MUST:
|
|
|
397
422
|
|
|
398
423
|
## FORBIDDEN Actions
|
|
399
424
|
|
|
400
|
-
| Action
|
|
401
|
-
|
|
|
402
|
-
|
|
|
403
|
-
| Skip typecheck
|
|
404
|
-
| Use `any` type
|
|
405
|
-
| Define types in `src/`
|
|
406
|
-
| Commit directly to main
|
|
407
|
-
| Skip documenter after implementation | Changelog + docs are mandatory
|
|
408
|
-
| Mix doc types in one file
|
|
409
|
-
| Leave docs unlinked from index
|
|
410
|
-
| Skip superpowers for features
|
|
411
|
-
| Skip code-simplifier
|
|
412
|
-
| Use MUI/Chakra
|
|
413
|
-
| Files > 400 lines
|
|
414
|
-
| 'use client' at top level
|
|
415
|
-
| Waterfall data fetching
|
|
416
|
-
| Skip CLAUDE.md update
|
|
425
|
+
| Action | Reason |
|
|
426
|
+
| ------------------------------------ | --------------------------------------------------------------------------------------------------------------------- |
|
|
427
|
+
| Speak/write in non-English | EN ONLY (chat, code, docs, commits) — see Language Policy. Override only via explicit user request, current turn only |
|
|
428
|
+
| Skip typecheck | Catches runtime errors |
|
|
429
|
+
| Use `any` type | Defeats strict mode |
|
|
430
|
+
| Define types in `src/` | Must be in `types/` |
|
|
431
|
+
| Commit directly to main | Create feature/fix branches |
|
|
432
|
+
| Skip documenter after implementation | Changelog + docs are mandatory |
|
|
433
|
+
| Mix doc types in one file | Changelog ≠ technical ≠ decision |
|
|
434
|
+
| Leave docs unlinked from index | Undiscoverable docs are useless |
|
|
435
|
+
| Skip superpowers for features | Use brainstorming + TDD |
|
|
436
|
+
| Skip code-simplifier | Run /simplify post-implementation |
|
|
437
|
+
| Use MUI/Chakra | Use shadcn/ui + Radix |
|
|
438
|
+
| Files > 400 lines | MUST split into smaller |
|
|
439
|
+
| 'use client' at top level | Push to leaf components only |
|
|
440
|
+
| Waterfall data fetching | Use Promise.all() for parallel |
|
|
441
|
+
| Skip CLAUDE.md update | MUST update after implementations |
|
|
417
442
|
|
|
418
443
|
---
|
|
419
444
|
|