specsmd 0.0.1 → 0.0.3

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 CHANGED
@@ -9,61 +9,15 @@ specsmd implements the [AI-Driven Development Lifecycle (AI-DLC)](https://aws.am
9
9
 
10
10
  ---
11
11
 
12
- ## What is AI-DLC?
13
-
14
- AI-DLC is a reimagined software development methodology where **AI drives the conversation** and humans validate. Unlike traditional Agile where iterations span weeks, AI-DLC operates in **Bolts** - rapid iterations measured in hours or days.
15
-
16
- > "Traditional development methods were built for human-driven, long-running processes. AI-DLC reimagines the development lifecycle with AI as a central collaborator, enabling rapid cycles measured in hours or days rather than weeks."
17
-
18
- ### AI-DLC vs Traditional Methods
19
-
20
- | Aspect | Agile/Scrum | AI-DLC |
21
- |--------|-------------|--------|
22
- | Iteration duration | Weeks (Sprints) | Hours/days (Bolts) |
23
- | Who drives | Human-driven, AI assists | AI-driven, human-validated |
24
- | Design techniques | Out of scope | Integrated (DDD, TDD, BDD) |
25
- | Task decomposition | Manual | AI-powered |
26
- | Phases | Repeating sprints | Rapid three-phase cycles (Inception → Construction → Operations) |
27
- | Rituals | Daily standups, retrospectives | Mob Elaboration, Mob Construction |
28
-
29
- ---
30
-
31
- ## How It Works
32
-
33
- specsmd provides four specialized agents that guide you through the entire development lifecycle:
34
-
35
- ```
36
- ┌─────────────────┐
37
- │ Master Agent │ Orchestrates & navigates
38
- └────────┬────────┘
39
-
40
- ┌────────────────────┼────────────────────┐
41
- ▼ ▼ ▼
42
- ┌───────────────┐ ┌───────────────┐ ┌───────────────┐
43
- │ Inception │ → │ Construction │ → │ Operations │
44
- │ Agent │ │ Agent │ │ Agent │
45
- └───────────────┘ └───────────────┘ └───────────────┘
46
- Capture intent Execute bolts Deploy & monitor
47
- Define units Build & test Verify & scale
48
- Plan stories Validate stages
49
- ```
50
-
51
- ### The Three Phases
52
-
53
- | Phase | Agent | Purpose | Key Outputs |
54
- |-------|-------|---------|-------------|
55
- | **Inception** | Inception Agent | Capture intents, elaborate requirements, decompose into units | User stories, NFRs, Unit definitions, Bolt plans |
56
- | **Construction** | Construction Agent | Execute bolts through domain design → logical design → code → test | Domain models, Technical designs, Code, Tests |
57
- | **Operations** | Operations Agent | Deploy, verify, and monitor | Deployment units, Monitoring, Runbooks |
58
-
59
- ---
60
-
61
12
  ## Quick Start
62
13
 
63
14
  ### Installation
64
15
 
16
+ > [!NOTE]
17
+ > Do not use npm if you want to always get the latest version. Use the npx command below.
18
+
65
19
  ```bash
66
- npx specsmd install
20
+ npx specsmd@latest install
67
21
  ```
68
22
 
69
23
  The installer detects your AI coding tools (Claude Code, Cursor, GitHub Copilot) and sets up:
@@ -72,6 +26,7 @@ The installer detects your AI coding tools (Claude Code, Cursor, GitHub Copilot)
72
26
  - Slash commands for easy agent invocation
73
27
 
74
28
  ### Initialize Your Project
29
+ Open your AI Asisted Tool (Claude Code, Cursor, GitHub Copilot) and run the following commands:
75
30
 
76
31
  ```bash
77
32
  # Start the Master Agent
@@ -126,6 +81,55 @@ Each bolt goes through validated stages:
126
81
 
127
82
  ---
128
83
 
84
+ ## What is AI-DLC?
85
+
86
+ AI-DLC is a reimagined software development methodology where **AI drives the conversation** and humans validate. Unlike traditional Agile where iterations span weeks, AI-DLC operates in **Bolts** - rapid iterations measured in hours or days.
87
+
88
+ > "Traditional development methods were built for human-driven, long-running processes. AI-DLC reimagines the development lifecycle with AI as a central collaborator, enabling rapid cycles measured in hours or days rather than weeks."
89
+
90
+ ### AI-DLC vs Traditional Methods
91
+
92
+ | Aspect | Agile/Scrum | AI-DLC |
93
+ |--------|-------------|--------|
94
+ | Iteration duration | Weeks (Sprints) | Hours/days (Bolts) |
95
+ | Who drives | Human-driven, AI assists | AI-driven, human-validated |
96
+ | Design techniques | Out of scope | Integrated (DDD, TDD, BDD) |
97
+ | Task decomposition | Manual | AI-powered |
98
+ | Phases | Repeating sprints | Rapid three-phase cycles (Inception → Construction → Operations) |
99
+ | Rituals | Daily standups, retrospectives | Mob Elaboration, Mob Construction |
100
+
101
+ ---
102
+
103
+ ## How It Works
104
+
105
+ specsmd provides four specialized agents that guide you through the entire development lifecycle:
106
+
107
+ ```
108
+ ┌─────────────────┐
109
+ │ Master Agent │ Orchestrates & navigates
110
+ └────────┬────────┘
111
+
112
+ ┌────────────────────┼────────────────────┐
113
+ ▼ ▼ ▼
114
+ ┌───────────────┐ ┌───────────────┐ ┌───────────────┐
115
+ │ Inception │ → │ Construction │ → │ Operations │
116
+ │ Agent │ │ Agent │ │ Agent │
117
+ └───────────────┘ └───────────────┘ └───────────────┘
118
+ Capture intent Execute bolts Deploy & monitor
119
+ Define units Build & test Verify & scale
120
+ Plan stories Validate stages
121
+ ```
122
+
123
+ ### The Three Phases
124
+
125
+ | Phase | Agent | Purpose | Key Outputs |
126
+ |-------|-------|---------|-------------|
127
+ | **Inception** | Inception Agent | Capture intents, elaborate requirements, decompose into units | User stories, NFRs, Unit definitions, Bolt plans |
128
+ | **Construction** | Construction Agent | Execute bolts through domain design → logical design → code → test | Domain models, Technical designs, Code, Tests |
129
+ | **Operations** | Operations Agent | Deploy, verify, and monitor | Deployment units, Monitoring, Runbooks |
130
+
131
+ ---
132
+
129
133
  ## Key Concepts
130
134
 
131
135
  ### Intent
@@ -62,7 +62,6 @@ Current types:
62
62
  - `ddd-construction-bolt` - Domain-Driven Design approach
63
63
 
64
64
 
65
-
66
65
  ---
67
66
 
68
67
  ## If Bolt Not Found
@@ -60,6 +60,7 @@ Review all stories to understand:
60
60
  **Read bolt type from unit-brief.md or use default based on unit type.**
61
61
 
62
62
  1. **Check unit-brief.md frontmatter** for `default_bolt_type`:
63
+
63
64
  ```yaml
64
65
  ---
65
66
  unit: expense-tracker-ui
@@ -77,6 +78,7 @@ Review all stories to understand:
77
78
  `.specsmd/aidlc/templates/construction/bolt-types/{bolt-type}.md`
78
79
 
79
80
  **Available bolt types**:
81
+
80
82
  - `ddd-construction-bolt` - For domain-heavy backend work (5 stages)
81
83
  - `simple-construction-bolt` - For UI, integrations, utilities (3 stages)
82
84
  - `spike-bolt` - For research and exploration
@@ -53,11 +53,13 @@ Break the Intent into independently deployable Units of Work based on project ty
53
53
  **CRITICAL**: Before decomposing, understand what types of units to create.
54
54
 
55
55
  1. **Read project type** from `memory-bank/project.yaml`:
56
+
56
57
  ```yaml
57
58
  project_type: full-stack-web # or backend-api, frontend-app, cli-tool, library
58
59
  ```
59
60
 
60
61
  2. **Read unit structure** from `catalog.yaml` under `project_types.{project_type}.unit_structure`:
62
+
61
63
  ```yaml
62
64
  unit_structure:
63
65
  backend:
@@ -217,6 +219,7 @@ Present proposed decomposition with their assigned requirements:
217
219
  Note: Story naming uses the story title (e.g., `001-user-signup.md`). No prefix field needed.
218
220
 
219
221
  **For frontend units**, also include:
222
+
220
223
  ```yaml
221
224
  unit_type: frontend
222
225
  default_bolt_type: simple-construction-bolt
@@ -11,7 +11,7 @@ class AntigravityInstaller extends ToolInstaller {
11
11
  }
12
12
 
13
13
  get commandsDir() {
14
- return path.join('.agent', 'agents');
14
+ return path.join('.agent', 'workflows');
15
15
  }
16
16
 
17
17
  get detectPath() {
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "specsmd",
3
- "version": "0.0.1",
3
+ "version": "0.0.3",
4
4
  "description": "Multi-agent orchestration system for AI-native software development. Delivers AI-DLC, Agile, and custom SDLC flows as markdown-based agent systems.",
5
5
  "main": "lib/installer.js",
6
6
  "bin": {