smash-os-install 0.3.3 → 0.3.5
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 +135 -0
- package/install.mjs +167 -334
- package/package.json +19 -3
- package/templates.mjs +2156 -0
package/README.md
ADDED
|
@@ -0,0 +1,135 @@
|
|
|
1
|
+
# smash-os-install
|
|
2
|
+
|
|
3
|
+
Install the SmashOS local AI workflow harness into any repository. No web app, no API keys, no cloud dependencies.
|
|
4
|
+
|
|
5
|
+
```bash
|
|
6
|
+
npx smash-os-install
|
|
7
|
+
```
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
|
|
11
|
+
## What SmashOS is
|
|
12
|
+
|
|
13
|
+
SmashOS gives Claude Code a coordinated team of 12 AI specialists — each with a defined role, scoring rules, and memory — that work together through structured pipelines to ship features, catch bugs, and audit security. Everything runs locally via Claude Code with no external services.
|
|
14
|
+
|
|
15
|
+
---
|
|
16
|
+
|
|
17
|
+
## Install
|
|
18
|
+
|
|
19
|
+
```bash
|
|
20
|
+
# Run inside your repo root
|
|
21
|
+
npx smash-os-install
|
|
22
|
+
```
|
|
23
|
+
|
|
24
|
+
**Prerequisites:** Node.js 18+ and [Claude Code](https://claude.ai/code)
|
|
25
|
+
|
|
26
|
+
The installer asks for your project name and tech stack, then writes all harness files into your repo.
|
|
27
|
+
|
|
28
|
+
---
|
|
29
|
+
|
|
30
|
+
## What gets created
|
|
31
|
+
|
|
32
|
+
```
|
|
33
|
+
your-repo/
|
|
34
|
+
├── CLAUDE.md ← harness instructions loaded by Claude on every session
|
|
35
|
+
├── .claude/
|
|
36
|
+
│ ├── hooks/
|
|
37
|
+
│ │ ├── boot-hook.mjs ← injects context files at session start
|
|
38
|
+
│ │ └── sync-hook.mjs ← persists memory after each response
|
|
39
|
+
│ └── agents/ ← 47 sub-agent definitions
|
|
40
|
+
├── ai/
|
|
41
|
+
│ ├── context/
|
|
42
|
+
│ │ ├── product.md ← fill in: what your product does
|
|
43
|
+
│ │ ├── architecture.md ← fill in: your stack and key decisions
|
|
44
|
+
│ │ └── coding-standards.md ← fill in: your style rules
|
|
45
|
+
│ ├── roles/ ← 12 role definitions (PM, Staff Eng, Dev, Security, QA, DevOps…)
|
|
46
|
+
│ ├── workflows/ ← 10 pipeline definitions
|
|
47
|
+
│ └── memory/
|
|
48
|
+
│ ├── decisions.md ← auto-populated: architectural decisions
|
|
49
|
+
│ ├── lessons.md ← auto-populated: what went wrong + why
|
|
50
|
+
│ └── constraints.md ← auto-populated: hard limits
|
|
51
|
+
└── .claude-state/ ← session breadcrumbs (gitignored)
|
|
52
|
+
├── active-context.md
|
|
53
|
+
├── next-steps.md
|
|
54
|
+
└── decisions.md
|
|
55
|
+
```
|
|
56
|
+
|
|
57
|
+
Skills install globally to `~/.claude/skills/` and are available in every project.
|
|
58
|
+
|
|
59
|
+
---
|
|
60
|
+
|
|
61
|
+
## First 3 commands
|
|
62
|
+
|
|
63
|
+
```bash
|
|
64
|
+
# 1. Open Claude Code in your repo
|
|
65
|
+
claude .
|
|
66
|
+
|
|
67
|
+
# 2. Verify the installation
|
|
68
|
+
/smash-os:status
|
|
69
|
+
|
|
70
|
+
# 3. Run your first pipeline
|
|
71
|
+
/smash-os:run feature "Add a health check endpoint"
|
|
72
|
+
```
|
|
73
|
+
|
|
74
|
+
---
|
|
75
|
+
|
|
76
|
+
## How a pipeline works
|
|
77
|
+
|
|
78
|
+
```
|
|
79
|
+
/smash-os:run feature "your task"
|
|
80
|
+
│
|
|
81
|
+
├─ Phase 1: Product Manager → spec + acceptance criteria (score ≥ 60 to pass)
|
|
82
|
+
├─ Phase 2: Staff Engineer → architecture review + spec sign-off
|
|
83
|
+
├─ Phase 3: Senior Developer → implementation (branch + commits)
|
|
84
|
+
├─ Phase 4: Security Engineer → threat model + security_score
|
|
85
|
+
├─ Phase 5: QA Engineer → test plan + test_success_score
|
|
86
|
+
└─ Phase 6: DevOps Engineer → PR + deployment readiness
|
|
87
|
+
```
|
|
88
|
+
|
|
89
|
+
Each phase scores its output 0–100. Below 60 halts the pipeline. The blocking reason is written to `ai/memory/` and injected into the next session automatically.
|
|
90
|
+
|
|
91
|
+
---
|
|
92
|
+
|
|
93
|
+
## Architecture
|
|
94
|
+
|
|
95
|
+
```
|
|
96
|
+
┌─────────────────────────────────────────────────────┐
|
|
97
|
+
│ Claude Code │
|
|
98
|
+
│ │
|
|
99
|
+
│ CLAUDE.md ──→ loads harness instructions │
|
|
100
|
+
│ boot-hook ──→ injects ai/context/ + ai/memory/ │
|
|
101
|
+
│ sync-hook ──→ writes decisions + lessons back │
|
|
102
|
+
│ │
|
|
103
|
+
│ /smash-os:run feature "task" │
|
|
104
|
+
│ │ │
|
|
105
|
+
│ └─→ Pipeline orchestrator │
|
|
106
|
+
│ │ │
|
|
107
|
+
│ ├─→ Role loader (ai/roles/) │
|
|
108
|
+
│ ├─→ Workflow runner (ai/workflows/) │
|
|
109
|
+
│ └─→ Sub-agents (.claude/agents/) │
|
|
110
|
+
│ │ │
|
|
111
|
+
│ └─→ Phase outputs → memory │
|
|
112
|
+
└─────────────────────────────────────────────────────┘
|
|
113
|
+
```
|
|
114
|
+
|
|
115
|
+
---
|
|
116
|
+
|
|
117
|
+
## Slash commands
|
|
118
|
+
|
|
119
|
+
| Command | Description |
|
|
120
|
+
|---------|-------------|
|
|
121
|
+
| `/smash-os:status` | Health check — active pipelines, recent decisions, memory state |
|
|
122
|
+
| `/smash-os:run [type]` | Start a pipeline: `feature`, `bug-fix`, `security-audit`, `weekly-improvement` |
|
|
123
|
+
| `/smash-os:fast` | Skip the pipeline — execute a trivial task inline (≤3 files) |
|
|
124
|
+
| `/smash-os:role [name]` | Switch to a specific role |
|
|
125
|
+
| `/smash-os:memory` | Show stored decisions, lessons, constraints |
|
|
126
|
+
| `/smash-os:audit` | Trigger an immediate security audit |
|
|
127
|
+
| `/smash-os:doctor` | Verify the installation is complete and correct |
|
|
128
|
+
| `/smash-os:evolve` | Run harness self-improvement on accumulated role learnings |
|
|
129
|
+
| `/smash-os:sync` | Manual mid-session checkpoint — save current state to memory |
|
|
130
|
+
|
|
131
|
+
---
|
|
132
|
+
|
|
133
|
+
## License
|
|
134
|
+
|
|
135
|
+
MIT
|