@spardutti/claude-skills 2.6.0 → 2.8.0
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 +16 -10
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,12 +1,22 @@
|
|
|
1
1
|
# Claude Skills
|
|
2
2
|
|
|
3
|
-
>
|
|
3
|
+
> **Make Claude Code write like a senior engineer who actually keeps up.**
|
|
4
4
|
|
|
5
5
|
[](https://www.npmjs.com/package/@spardutti/claude-skills)
|
|
6
6
|
[](https://www.npmjs.com/package/@spardutti/claude-skills)
|
|
7
7
|
[](./LICENSE)
|
|
8
8
|
|
|
9
|
-
|
|
9
|
+
Opinionated, always-current best-practice playbooks that load while Claude codes — so it writes today's idioms for the tools you actually use, not the averaged, two-years-stale code its training defaults to. This repo is the source catalog; the CLI installs exactly what each project needs from an interactive menu, pulling in any subagents your chosen commands depend on.
|
|
10
|
+
|
|
11
|
+
## Why This Exists
|
|
12
|
+
|
|
13
|
+
Coding agents don't fail because they can't code — they fail because they write **dated** code. A model's knowledge is frozen and averaged: ask for React and it reaches for `useEffect` and manual `useMemo`; ask for TypeScript and it assumes 5.x; it defaults to patterns that were fine two years ago and are wrong today.
|
|
14
|
+
|
|
15
|
+
The ecosystem's answer has been **process** — planning frameworks, spec-driven workflows, agent orchestration. Those steer *how* the agent works. None of them fix *what the code looks like*.
|
|
16
|
+
|
|
17
|
+
**This is the other half.** Each skill is a tight, opinionated playbook that pins Claude to current, senior-level practice for one tool — React 19.2 + the React Compiler, TypeScript 7.0, TanStack Query v5, Drizzle, FastAPI, SQL. Claude loads only the skills a task touches, so many can stay installed cheaply. Bring whatever planning framework you like on top; this makes sure the code that comes out isn't 2023-vintage.
|
|
18
|
+
|
|
19
|
+
**Freshness is the feature.** These track the changelogs as the tools move — the React Compiler going stable, TS 7.0's native compiler, Drizzle Relations v2 — so you don't have to.
|
|
10
20
|
|
|
11
21
|
## Quick Start
|
|
12
22
|
|
|
@@ -17,7 +27,7 @@ npx @spardutti/claude-skills
|
|
|
17
27
|
```
|
|
18
28
|
|
|
19
29
|
```text
|
|
20
|
-
Claude Skills Installer v2.
|
|
30
|
+
Claude Skills Installer v2.6.0
|
|
21
31
|
|
|
22
32
|
── Frontend ──────────────────────────────
|
|
23
33
|
◉ react ◯ tanstack-query
|
|
@@ -69,12 +79,6 @@ The CLI will:
|
|
|
69
79
|
|-------|----------------|
|
|
70
80
|
| `sql` 📦 | Schema design, data types, indexing & `EXPLAIN`, joins & subqueries, ORM patterns (N+1, transactions, locking), safe migrations |
|
|
71
81
|
|
|
72
|
-
### Desktop
|
|
73
|
-
|
|
74
|
-
| Skill | What it covers |
|
|
75
|
-
|-------|----------------|
|
|
76
|
-
| `tauri-v2` | Tauri v2 — IPC commands, plugins, window management, system tray, global shortcuts, capabilities/permissions, events |
|
|
77
|
-
|
|
78
82
|
### Foundations
|
|
79
83
|
|
|
80
84
|
Cross-cutting craft — applies to any stack, any language.
|
|
@@ -82,7 +86,8 @@ Cross-cutting craft — applies to any stack, any language.
|
|
|
82
86
|
| Skill | What it covers |
|
|
83
87
|
|-------|----------------|
|
|
84
88
|
| `code-structure` 📦 | Single Responsibility (when to split) + Avoid Hasty Abstractions (when *not* to extract) — hard size limits, separation of concerns, the Rule of Three |
|
|
85
|
-
| `
|
|
89
|
+
| `debugging` | Root cause before patching — reproduce, trace the failure backward to its origin, one hypothesis at a time, the 3-attempts-then-question-the-design rule |
|
|
90
|
+
| `typescript-best-practices` | TypeScript 7.x — type design, generics, type guards, `satisfies`, `using`, error handling, `tsconfig` |
|
|
86
91
|
| `testing-best-practices` | Arrange-Act-Assert, factory-based test data, isolation, mocking boundaries, a pyramid-balanced suite |
|
|
87
92
|
| `security-practices` | OWASP Top 10 prevention, input validation, auth, SQL injection, XSS, CSRF, secure defaults |
|
|
88
93
|
|
|
@@ -93,6 +98,7 @@ Portable slash commands installed to `.claude/commands/`. Some orchestrate paral
|
|
|
93
98
|
| Command | What it does |
|
|
94
99
|
|---------|--------------|
|
|
95
100
|
| `/ship` | Unified delivery pipeline — commit → PR → merge → release. No argument steps through interactively; `/ship pr` runs through PR creation; `/ship release` runs the full pipeline |
|
|
101
|
+
| `/discover` | Find the right problem before deciding what to build — diverges first: names competing interpretations of what you said, stress-tests the framing with a blind critic subagent, then converges on a mental model. Run before `/preplan` |
|
|
96
102
|
| `/preplan` | Resolve a fuzzy feature idea into concrete decisions — 6 fixed phases, one question at a time, ends with a decision log. Run before `/plan-feature` |
|
|
97
103
|
| `/plan-feature` | Integration-first feature planning — 3 parallel subagents scan for reusable code, patterns, and touch points before producing a short plan |
|
|
98
104
|
| `/refactor` | Detect size / complexity / duplication / coupling issues via 4 parallel subagents, then refactor |
|