@kennethsolomon/shipkit 3.18.0 → 3.20.0
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.
- package/README.md +25 -3
- package/package.json +1 -1
- package/skills/sk:brainstorming/SKILL.md +19 -128
- package/skills/sk:debug/SKILL.md +44 -111
- package/skills/sk:e2e/SKILL.md +45 -97
- package/skills/sk:features/SKILL.md +44 -99
- package/skills/sk:frontend-design/SKILL.md +16 -32
- package/skills/sk:laravel-deploy/SKILL.md +112 -0
- package/skills/sk:laravel-init/SKILL.md +18 -0
- package/skills/sk:lint/SKILL.md +42 -62
- package/skills/sk:mvp/SKILL.md +81 -134
- package/skills/sk:perf/SKILL.md +24 -43
- package/skills/sk:review/SKILL.md +57 -93
- package/skills/sk:security-check/SKILL.md +37 -43
- package/skills/sk:seo-audit/SKILL.md +75 -96
- package/skills/sk:setup-claude/SKILL.md +148 -0
- package/skills/sk:setup-claude/references/skill-profiles.md +201 -0
- package/skills/sk:setup-claude/templates/.claude/agents/debugger.md +7 -0
- package/skills/sk:setup-claude/templates/.claude/agents/qa-engineer.md +8 -0
- package/skills/sk:setup-claude/templates/.claude/agents/tech-writer.md +7 -0
- package/skills/sk:setup-claude/templates/.claude/rules/laravel.md.template +11 -0
- package/skills/sk:setup-claude/templates/CLAUDE.md.template +102 -244
- package/skills/sk:setup-claude/templates/commands/brainstorm.md.template +1 -1
- package/skills/sk:setup-claude/templates/commands/execute-plan.md.template +1 -1
- package/skills/sk:setup-claude/templates/commands/finish-feature.md.template +1 -1
- package/skills/sk:setup-claude/templates/commands/security-check.md.template +1 -1
- package/skills/sk:setup-claude/templates/commands/write-plan.md.template +1 -1
- package/skills/sk:setup-optimizer/SKILL.md +85 -14
- package/skills/sk:skill-creator/SKILL.md +115 -226
- package/skills/sk:website/SKILL.md +81 -149
- package/skills/sk:write-tests/SKILL.md +44 -110
|
@@ -0,0 +1,201 @@
|
|
|
1
|
+
# ShipKit Skill Profiles — Stack-Based Filtering
|
|
2
|
+
|
|
3
|
+
**Last Updated:** 2026-03-29
|
|
4
|
+
**Total Skills:** 44 | **Agents:** 13 | **Rules:** 6
|
|
5
|
+
|
|
6
|
+
This file is the source of truth for which skills, agents, and rules get installed per project. Read by `sk:setup-claude` (initial setup) and `sk:setup-optimizer` (ongoing sync).
|
|
7
|
+
|
|
8
|
+
---
|
|
9
|
+
|
|
10
|
+
## Universal Skills (always installed — 25 skills)
|
|
11
|
+
|
|
12
|
+
| Skill | Purpose |
|
|
13
|
+
|-------|---------|
|
|
14
|
+
| sk:start | Smart entry point — classifies task, routes to workflow |
|
|
15
|
+
| sk:context | Load project context + session brief |
|
|
16
|
+
| sk:brainstorming | Explore requirements before implementation |
|
|
17
|
+
| sk:write-plan | Decision-complete plan into todo.md |
|
|
18
|
+
| sk:execute-plan | Implement tasks in batches with progress logging |
|
|
19
|
+
| sk:write-tests | TDD: write failing tests before implementation |
|
|
20
|
+
| sk:test | Run all test suites, verify 100% coverage |
|
|
21
|
+
| sk:lint | Auto-detect and run all linters |
|
|
22
|
+
| sk:debug | Structured bug investigation |
|
|
23
|
+
| sk:review | 7-dimension self-review |
|
|
24
|
+
| sk:smart-commit | Conventional commit with approval |
|
|
25
|
+
| sk:gates | All quality gates in parallel batches |
|
|
26
|
+
| sk:safety-guard | Protect against destructive operations |
|
|
27
|
+
| sk:scope-check | Detect scope creep vs plan |
|
|
28
|
+
| sk:save-session | Save session state for continuity |
|
|
29
|
+
| sk:resume-session | Restore previously saved session |
|
|
30
|
+
| sk:learn | Extract reusable patterns from session |
|
|
31
|
+
| sk:retro | Post-ship retrospective |
|
|
32
|
+
| sk:health | Harness self-audit scorecard |
|
|
33
|
+
| sk:context-budget | Audit context window token consumption |
|
|
34
|
+
| sk:setup-claude | Bootstrap project scaffolding |
|
|
35
|
+
| sk:setup-optimizer | Diagnose + update workflow + enrich CLAUDE.md |
|
|
36
|
+
| sk:release | Version bump + changelog + tag |
|
|
37
|
+
| sk:fast-track | Abbreviated workflow for small changes |
|
|
38
|
+
| sk:skill-creator | Create or modify skills |
|
|
39
|
+
|
|
40
|
+
---
|
|
41
|
+
|
|
42
|
+
## Stack-Specific Skills
|
|
43
|
+
|
|
44
|
+
### Laravel (+4 skills)
|
|
45
|
+
|
|
46
|
+
| Skill | Purpose |
|
|
47
|
+
|-------|---------|
|
|
48
|
+
| sk:laravel-new | Scaffold fresh Laravel app |
|
|
49
|
+
| sk:laravel-init | Configure existing Laravel project |
|
|
50
|
+
| sk:laravel-deploy | Deploy to Laravel Cloud |
|
|
51
|
+
| sk:schema-migrate | Multi-ORM schema change analysis |
|
|
52
|
+
|
|
53
|
+
### Web (+5 skills)
|
|
54
|
+
|
|
55
|
+
| Skill | Purpose |
|
|
56
|
+
|-------|---------|
|
|
57
|
+
| sk:frontend-design | UI mockup before implementation |
|
|
58
|
+
| sk:accessibility | WCAG 2.1 AA audit |
|
|
59
|
+
| sk:seo-audit | SEO audit (source + dev server) |
|
|
60
|
+
| sk:website | Build multi-page marketing website |
|
|
61
|
+
| sk:mvp | Generate MVP with landing page + app |
|
|
62
|
+
|
|
63
|
+
### Database (+1 skill)
|
|
64
|
+
|
|
65
|
+
| Skill | Purpose |
|
|
66
|
+
|-------|---------|
|
|
67
|
+
| sk:schema-migrate | Multi-ORM schema change analysis |
|
|
68
|
+
|
|
69
|
+
### API (+2 skills)
|
|
70
|
+
|
|
71
|
+
| Skill | Purpose |
|
|
72
|
+
|-------|---------|
|
|
73
|
+
| sk:api-design | Design API contracts before implementation |
|
|
74
|
+
| sk:team | Parallel domain agents (BE + FE + QA) |
|
|
75
|
+
|
|
76
|
+
### Mobile (exclude from web)
|
|
77
|
+
|
|
78
|
+
No additional skills. **Exclude:** sk:e2e (Playwright), sk:seo-audit, sk:accessibility, sk:website.
|
|
79
|
+
|
|
80
|
+
---
|
|
81
|
+
|
|
82
|
+
## Opt-In Skills (manual activation only — 8 skills)
|
|
83
|
+
|
|
84
|
+
| Skill | Purpose | When to activate |
|
|
85
|
+
|-------|---------|-----------------|
|
|
86
|
+
| sk:autopilot | Hands-free workflow, all 8 steps | Well-defined tasks, minimal steering |
|
|
87
|
+
| sk:eval | Run evaluations for agent reliability | Eval-driven development |
|
|
88
|
+
| sk:ci | Set up GitHub Actions or GitLab CI | First-time CI setup |
|
|
89
|
+
| sk:reverse-doc | Generate docs from existing code | Onboarding, formalizing prototypes |
|
|
90
|
+
| sk:features | Sync feature specs with implementation | Feature spec maintenance |
|
|
91
|
+
| sk:e2e | E2E behavioral verification | Web projects with Playwright/Cypress |
|
|
92
|
+
| sk:dashboard | Workflow Kanban board (localhost) | Multi-worktree progress tracking |
|
|
93
|
+
| sk:plugin | Package skills as distributable plugin | Team skill sharing |
|
|
94
|
+
|
|
95
|
+
---
|
|
96
|
+
|
|
97
|
+
## Quality Gate Skills (part of sk:gates — always installed)
|
|
98
|
+
|
|
99
|
+
| Skill | Gate role |
|
|
100
|
+
|-------|----------|
|
|
101
|
+
| sk:perf | Performance audit |
|
|
102
|
+
| sk:security-check | OWASP security audit |
|
|
103
|
+
|
|
104
|
+
These are installed universally because they run inside `sk:gates`. Perf auto-skips if no frontend+DB keywords. Security always runs.
|
|
105
|
+
|
|
106
|
+
---
|
|
107
|
+
|
|
108
|
+
## Stack Detection Rules
|
|
109
|
+
|
|
110
|
+
| Priority | Signal | Detected Stack | Capabilities |
|
|
111
|
+
|----------|--------|---------------|-------------|
|
|
112
|
+
| 1 | `composer.json` + `laravel/framework` | laravel | web, database, api |
|
|
113
|
+
| 2 | `package.json` + `next` | nextjs | web |
|
|
114
|
+
| 3 | `package.json` + `nuxt` | nuxt | web |
|
|
115
|
+
| 4 | `package.json` + `react` (no next) | react | web |
|
|
116
|
+
| 5 | `package.json` + `vue` (no nuxt) | vue | web |
|
|
117
|
+
| 6 | `package.json` + `svelte` | svelte | web |
|
|
118
|
+
| 7 | `app.json` or `app.config.ts` | expo | mobile |
|
|
119
|
+
| 8 | `react-native.config.js` | react-native | mobile |
|
|
120
|
+
| 9 | `pubspec.yaml` | flutter | mobile |
|
|
121
|
+
| 10 | `package.json` + `express` | express | api |
|
|
122
|
+
| 11 | `go.mod` | go | api |
|
|
123
|
+
| 12 | `Cargo.toml` | rust | api |
|
|
124
|
+
| 13 | `pyproject.toml` / `requirements.txt` | python | api |
|
|
125
|
+
| 14 | `Gemfile` + `rails` | rails | web, database, api |
|
|
126
|
+
|
|
127
|
+
### Capability → Add-on mapping
|
|
128
|
+
|
|
129
|
+
| Capability | Add-on skills |
|
|
130
|
+
|-----------|---------------|
|
|
131
|
+
| web | sk:frontend-design, sk:accessibility, sk:seo-audit, sk:website, sk:mvp |
|
|
132
|
+
| database | sk:schema-migrate |
|
|
133
|
+
| api | sk:api-design, sk:team |
|
|
134
|
+
| laravel | sk:laravel-new, sk:laravel-init, sk:laravel-deploy |
|
|
135
|
+
| mobile | Exclude: sk:e2e, sk:seo-audit, sk:accessibility, sk:website |
|
|
136
|
+
|
|
137
|
+
### Database sub-detection (within any stack)
|
|
138
|
+
|
|
139
|
+
| Signal | ORM |
|
|
140
|
+
|--------|-----|
|
|
141
|
+
| `prisma/schema.prisma` | Prisma → add `database` capability |
|
|
142
|
+
| `drizzle.config.ts` / `.js` | Drizzle → add `database` capability |
|
|
143
|
+
| `database/migrations/` (Laravel) | Laravel migrations → add `database` capability |
|
|
144
|
+
| `alembic/` | SQLAlchemy → add `database` capability |
|
|
145
|
+
| `db/migrate/` (Rails) | Rails migrations → add `database` capability |
|
|
146
|
+
|
|
147
|
+
---
|
|
148
|
+
|
|
149
|
+
## Agent → Stack Mapping
|
|
150
|
+
|
|
151
|
+
| Agent | Stacks | Notes |
|
|
152
|
+
|-------|--------|-------|
|
|
153
|
+
| architect | all | System design, universal |
|
|
154
|
+
| backend-dev | laravel, express, go, python, rust, rails | Backend implementation |
|
|
155
|
+
| frontend-dev | react, nextjs, vue, nuxt, svelte | Frontend implementation |
|
|
156
|
+
| mobile-dev | expo, react-native, flutter | Mobile-specific |
|
|
157
|
+
| database-architect | any with `database` capability | Schema design, migrations |
|
|
158
|
+
| qa-engineer | all | E2E test scenarios, universal |
|
|
159
|
+
| debugger | all | Bug investigation, universal |
|
|
160
|
+
| code-reviewer | all | Code quality review, universal |
|
|
161
|
+
| security-reviewer | all | OWASP audit, universal |
|
|
162
|
+
| performance-optimizer | all | Performance analysis, universal |
|
|
163
|
+
| refactor-specialist | all | Safe refactoring, universal |
|
|
164
|
+
| tech-writer | all | Documentation, universal |
|
|
165
|
+
| devops-engineer | all | CI/CD, deployment, universal |
|
|
166
|
+
|
|
167
|
+
---
|
|
168
|
+
|
|
169
|
+
## Rule → Stack Mapping
|
|
170
|
+
|
|
171
|
+
| Rule | Applies to stacks | Path patterns |
|
|
172
|
+
|------|------------------|---------------|
|
|
173
|
+
| tests.md | all | `tests/**`, `**/*.test.*`, `**/*.spec.*` |
|
|
174
|
+
| api.md | laravel, express, go, python, rails | `routes/api.php`, `**/controllers/**`, `src/api/**` |
|
|
175
|
+
| laravel.md | laravel | `app/**/*.php`, `routes/**`, `config/**` |
|
|
176
|
+
| react.md | react, nextjs | `**/*.{tsx,jsx}`, `resources/js/**` |
|
|
177
|
+
| vue.md | vue, nuxt | `**/*.vue`, `resources/js/**` |
|
|
178
|
+
| migrations.md | any with `database` capability | `database/migrations/**`, `prisma/**`, `db/migrate/**` |
|
|
179
|
+
|
|
180
|
+
---
|
|
181
|
+
|
|
182
|
+
## Installation Formula
|
|
183
|
+
|
|
184
|
+
For a given project with detected `stack` and `capabilities`:
|
|
185
|
+
|
|
186
|
+
```
|
|
187
|
+
installed_skills = universal_skills
|
|
188
|
+
+ capability_add_ons(capabilities)
|
|
189
|
+
+ config.skills.extra
|
|
190
|
+
- config.skills.disabled
|
|
191
|
+
- mobile_exclusions (if mobile stack)
|
|
192
|
+
|
|
193
|
+
installed_agents = universal_agents
|
|
194
|
+
+ stack_specific_agents(stack)
|
|
195
|
+
- agents not matching any detected stack
|
|
196
|
+
|
|
197
|
+
installed_rules = universal_rules (tests.md)
|
|
198
|
+
+ stack_matching_rules(stack, capabilities)
|
|
199
|
+
```
|
|
200
|
+
|
|
201
|
+
User overrides (`extra`, `disabled`) are never touched by auto-detection.
|
|
@@ -6,6 +6,13 @@ allowed-tools: Read, Edit, Bash, Grep, Glob
|
|
|
6
6
|
memory: project
|
|
7
7
|
---
|
|
8
8
|
|
|
9
|
+
<!-- DESIGN NOTE: No `isolation: worktree` by design.
|
|
10
|
+
Debugger is invoked solo (never in parallel with other agents) and must
|
|
11
|
+
see the actual working state — the real failing code, the real test output,
|
|
12
|
+
the real stack trace. A worktree copy would give it a clean slate,
|
|
13
|
+
defeating its ability to reproduce the bug. Edit access is needed to place
|
|
14
|
+
and remove targeted debug logs during investigation. -->
|
|
15
|
+
|
|
9
16
|
# Debugger Agent
|
|
10
17
|
|
|
11
18
|
You are an expert debugger. Find root causes, not symptoms.
|
|
@@ -7,6 +7,14 @@ memory: project
|
|
|
7
7
|
background: true
|
|
8
8
|
---
|
|
9
9
|
|
|
10
|
+
<!-- DESIGN NOTE: No `isolation: worktree` by design.
|
|
11
|
+
qa-engineer runs as a background agent (background: true) alongside
|
|
12
|
+
backend-dev and frontend-dev during sk:team. It only creates test scenario
|
|
13
|
+
files (e.g., e2e/<feature>.spec.ts, tasks/e2e-scenarios.md) — never source
|
|
14
|
+
files. Its writes land in a separate directory tree from the implementation
|
|
15
|
+
agents' worktree changes, so file conflicts cannot occur. Isolation adds
|
|
16
|
+
overhead with no benefit here. -->
|
|
17
|
+
|
|
10
18
|
# QA Engineer Agent
|
|
11
19
|
|
|
12
20
|
You are the QA Agent in a team workflow. Your job is to write E2E test scenarios while the Backend and Frontend agents implement code in parallel.
|
|
@@ -6,6 +6,13 @@ tools: Read, Write, Edit, Grep, Glob
|
|
|
6
6
|
memory: project
|
|
7
7
|
---
|
|
8
8
|
|
|
9
|
+
<!-- DESIGN NOTE: No `isolation: worktree` by design.
|
|
10
|
+
tech-writer only creates and edits documentation files (README.md,
|
|
11
|
+
docs/*.md, architecture docs, inline comments). It never modifies source
|
|
12
|
+
code, migrations, or config files. Because its writes are confined to
|
|
13
|
+
documentation paths that implementation agents never touch, isolation is
|
|
14
|
+
unnecessary overhead. -->
|
|
15
|
+
|
|
9
16
|
You are a technical writer specializing in developer documentation. You make codebases comprehensible — to future contributors, to users, and to the developers themselves six months later.
|
|
10
17
|
|
|
11
18
|
## On Invocation
|
|
@@ -18,3 +18,14 @@ paths:
|
|
|
18
18
|
- **Migrations**: One logical change per migration. Never modify a published migration.
|
|
19
19
|
- **Config**: Access config via `config()` helper, never `env()` outside config files.
|
|
20
20
|
- **Strict mode**: Models use strict mode (prevent lazy loading, silently discarding attributes, accessing missing attributes).
|
|
21
|
+
|
|
22
|
+
## Code Refinement
|
|
23
|
+
|
|
24
|
+
After running `/sk:execute-plan`, use the `laravel-simplifier` agent to refine recently modified PHP/Laravel code:
|
|
25
|
+
- Applies PSR-12 standards and Laravel conventions
|
|
26
|
+
- Reduces unnecessary nesting and complexity — prefer `match` over nested ternaries
|
|
27
|
+
- Improves variable and method naming for readability
|
|
28
|
+
- Never changes behavior — refine only
|
|
29
|
+
|
|
30
|
+
Install once: `/plugin install laravel-simplifier@laravel`
|
|
31
|
+
Invoke: "Review recent changes using the laravel-simplifier agent"
|