@codihaus/claude-skills 1.6.26 → 1.6.27

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 CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@codihaus/claude-skills",
3
- "version": "1.6.26",
3
+ "version": "1.6.27",
4
4
  "description": "Claude Code skills for software development workflow",
5
5
  "main": "src/index.js",
6
6
  "bin": {
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  name: debrief
3
3
  description: Customer requirements → questionnaire (default) or BRD (with --answers flag)
4
- version: 5.2.0
4
+ version: 5.3.0
5
5
  ---
6
6
 
7
7
  # /debrief - Business Requirements Document
@@ -91,7 +91,13 @@ version: 5.2.0
91
91
 
92
92
  ### 4. Questionnaire = Decision Tool
93
93
 
94
- **3 sheets with evidence**:
94
+ **How to generate**:
95
+ ```bash
96
+ python .claude/skills/debrief/scripts/generate_questionnaire.py questionnaire-{YYYY-MM-DD}.xlsx -
97
+ ```
98
+ Pipe JSON to stdin with research data (see `references/workflow.md` for JSON format).
99
+
100
+ **3 sheets generated**:
95
101
  - Summary (features count, tier breakdown, research stats)
96
102
  - Questions (validation + open)
97
103
  - References (URLs, evidence per feature)
@@ -106,13 +112,24 @@ Then create BRD via `--answers` or `--generate-brd`.
106
112
 
107
113
  **See** `references/workflow.md` for detailed steps.
108
114
 
109
- **Modes**:
110
- - New Project: Research + questionnaire (no BRD)
111
- - Add Feature: Research + questionnaire (no BRD)
112
- - Process Answers (`--answers`): Create BRD from filled questionnaire
113
- - Generate BRD (`--generate-brd`): Create BRD from research
115
+ **CRITICAL**: ALWAYS ask context questions FIRST using AskUserQuestion tool before doing any research.
114
116
 
115
- **Default**: Questionnaire only. BRD requires explicit flag.
117
+ **Modes**:
118
+ - **New Project**:
119
+ 1. Ask 4 questions (Industry, Users, Constraints, Scope mode: MVP/Market Standard/Full)
120
+ 2. Research + questionnaire → `plans/brd/use-cases/{project}/questionnaire-{date}.xlsx`
121
+ 3. NO BRD created
122
+ - **Add Feature**:
123
+ 1. Ask 2 questions (Feature name, Scope mode: MVP/Market Standard/Full)
124
+ 2. Research + questionnaire → `plans/brd/use-cases/{feature}/questionnaire-{date}.xlsx`
125
+ 3. NO BRD created
126
+ - **Process Answers** (`--answers`):
127
+ 1. Read filled questionnaire
128
+ 2. **ASK what to do**: Continue research / Ask follow-ups / Create BRD
129
+ 3. Execute based on choice
130
+ - **Generate BRD** (`--generate-brd`): Create BRD from research
131
+
132
+ **Default**: Questionnaire only. BRD requires explicit flag AND user confirmation.
116
133
 
117
134
  ---
118
135
 
@@ -28,26 +28,30 @@
28
28
 
29
29
  **Mindset**: Discovery mode. Understand context, research market, validate with customer.
30
30
 
31
- ### Context Gathering
31
+ ### STEP 1: Context Gathering (MANDATORY - DO THIS FIRST)
32
+
33
+ **CRITICAL**: ALWAYS ask questions BEFORE doing any research. DO NOT skip this step.
32
34
 
33
35
  **Check for existing questionnaire**:
34
36
  - If found: Ask "Found existing research. Use it or start fresh?"
35
37
  - Reuse → Switch to Generate BRD mode
36
38
  - Fresh → Continue with new research
37
39
 
38
- **Ask 4 questions** (AskUserQuestion):
39
- 1. Industry (SaaS/E-commerce/Marketplace/etc.)
40
- 2. Target users (B2B/B2C/Internal)
41
- 3. Constraints (timeline/budget/compliance/integrations)
42
- 4. **Scope tier** (Core/Standard/Full) - determines WHICH features to include
40
+ **ALWAYS ask 4 questions** using AskUserQuestion tool:
41
+ 1. **Industry**: SaaS / E-commerce / Marketplace / etc.
42
+ 2. **Target users**: B2B / B2C / Internal
43
+ 3. **Constraints**: Timeline / Budget / Compliance / Integrations
44
+ 4. **Scope mode**: MVP / Market Standard / Full (determines WHICH features to include)
45
+
46
+ **DO NOT proceed to research until user answers these questions.**
43
47
 
44
48
  **Output**: Context understanding, scope alignment
45
49
 
46
50
  ---
47
51
 
48
- ### Market Research (ALWAYS Deep + Revalidation)
52
+ ### STEP 2: Market Research (ALWAYS Deep + Revalidation)
49
53
 
50
- **CRITICAL**: After scope tier chosen, ALWAYS do deep research + multi-source revalidation.
54
+ **CRITICAL**: After user answers questions and scope tier is chosen, THEN do deep research + multi-source revalidation.
51
55
 
52
56
  **Methodology**: See `research.md` for full process.
53
57
 
@@ -65,16 +69,52 @@
65
69
 
66
70
  ---
67
71
 
68
- ### Generate Questionnaire (ALWAYS)
69
-
70
- **Principle**: Default output. No BRD files created yet.
71
-
72
- **3 sheets**:
72
+ ### STEP 3: Generate Questionnaire (ALWAYS)
73
+
74
+ **Principle**: After research is complete, generate questionnaire. No BRD files created yet.
75
+
76
+ **How to generate**:
77
+ 1. Create JSON with research data (see format below)
78
+ 2. Call Python script: `python .claude/skills/debrief/scripts/generate_questionnaire.py questionnaire-{YYYY-MM-DD}.xlsx -`
79
+ 3. Pipe JSON to stdin or save as file
80
+
81
+ **JSON format**:
82
+ ```json
83
+ {
84
+ "project_name": "Project Name",
85
+ "date": "2026-01-25",
86
+ "research": {
87
+ "depth": "Deep",
88
+ "features_count": 12,
89
+ "tiers": {"MVP": 5, "Standard": 4, "Advanced": 3},
90
+ "sources_count": 23
91
+ },
92
+ "references": {
93
+ "Feature Name": [
94
+ {"type": "Comparison", "url": "https://..."},
95
+ {"type": "G2 Reviews", "url": "https://..."}
96
+ ]
97
+ },
98
+ "questions": [
99
+ {
100
+ "category": "Requirements",
101
+ "question": "Question text?",
102
+ "priority": "Required",
103
+ "context": "Why we're asking",
104
+ "source": "UC-XXX-NNN"
105
+ }
106
+ ]
107
+ }
108
+ ```
109
+
110
+ **3 sheets generated**:
73
111
  - **Summary**: Feature count, tier distribution, research stats
74
112
  - **Questions**: Validation + open questions
75
113
  - **References**: URLs by feature (comparison pages, reviews, ecosystems)
76
114
 
77
- **Output**: `questionnaire-{date}.xlsx` in current directory
115
+ **Output location**:
116
+ - New Project: `plans/brd/use-cases/{project-name}/questionnaire-{YYYY-MM-DD}.xlsx`
117
+ - Add Feature: `plans/brd/use-cases/{feature-name}/questionnaire-{YYYY-MM-DD}.xlsx`
78
118
 
79
119
  **NO BRD files created**. User reviews questionnaire first.
80
120
 
@@ -101,25 +141,29 @@
101
141
 
102
142
  **Mindset**: Continuity mode. Check duplicates, maintain consistency.
103
143
 
104
- ### Context Gathering
144
+ ### STEP 1: Context Gathering (MANDATORY - DO THIS FIRST)
145
+
146
+ **CRITICAL**: ALWAYS ask questions BEFORE doing any research. DO NOT skip this step.
105
147
 
106
- **Ask 2 questions**:
107
- 1. Feature name
108
- 2. Scope tier (Core/Standard/Full)
148
+ **ALWAYS ask 2 questions** using AskUserQuestion tool:
149
+ 1. **Feature name**: What feature to add?
150
+ 2. **Scope mode**: MVP / Market Standard / Full (determines WHICH features to include)
109
151
 
110
152
  **Duplicate check**: Read `docs-graph.json` if exists, warn if similar feature found
111
153
 
154
+ **DO NOT proceed to research until user answers these questions.**
155
+
112
156
  **Output**: Feature to add, scope confirmed
113
157
 
114
158
  ---
115
159
 
116
- ### Research & Generate Questionnaire
160
+ ### STEP 2: Research & Generate Questionnaire
117
161
 
118
162
  **Same as New Project**:
119
163
  - Market research (comparison-first, see `research.md`)
120
164
  - Generate questionnaire (validation + open questions)
121
165
 
122
- **Output**: `features/{feature}/questionnaire-{date}.xlsx`
166
+ **Output**: `plans/brd/use-cases/{feature}/questionnaire-{YYYY-MM-DD}.xlsx`
123
167
 
124
168
  **NO BRD files created**. User reviews questionnaire, gets answers, then runs `--answers`.
125
169
 
@@ -129,11 +173,11 @@
129
173
 
130
174
  **Trigger**: `/debrief --answers questionnaire.xlsx`
131
175
 
132
- **Goal**: Create BRD from customer-answered questionnaire.
176
+ **Goal**: Process filled questionnaire and ask what user wants to do next.
133
177
 
134
- **Mindset**: BRD creation mode. Customer validated research, now create structure.
178
+ **Mindset**: Interactive mode. User decides next action based on answers.
135
179
 
136
- ### Read Questionnaire
180
+ ### STEP 1: Read Questionnaire
137
181
 
138
182
  **Read questionnaire** (Excel file):
139
183
  - Summary sheet (features, tiers, research)
@@ -142,7 +186,23 @@
142
186
 
143
187
  ---
144
188
 
145
- ### Feature Sequencing
189
+ ### STEP 2: Ask What To Do (MANDATORY)
190
+
191
+ **CRITICAL**: ALWAYS ask what user wants to do. DO NOT auto-create BRD.
192
+
193
+ **Ask using AskUserQuestion tool** - What would you like to do?
194
+ 1. **Continue research and revise summary** (do more research based on answers)
195
+ 2. **Ask follow-up questions** (generate new questionnaire with deeper questions)
196
+ 3. **Create BRD** (generate use cases and BRD files)
197
+
198
+ **Based on user choice**:
199
+ - Choice 1 → Do additional research, update questionnaire
200
+ - Choice 2 → Generate new questionnaire with follow-up questions
201
+ - Choice 3 → Proceed to Feature Sequencing (below)
202
+
203
+ ---
204
+
205
+ ### STEP 3: Feature Sequencing (ONLY if user chose "Create BRD")
146
206
 
147
207
  **Principle**: Order by dependencies within each tier.
148
208
 
@@ -159,7 +219,7 @@
159
219
 
160
220
  ---
161
221
 
162
- ### Use Case Generation
222
+ ### STEP 4: Use Case Generation (ONLY if user chose "Create BRD")
163
223
 
164
224
  **Principle**: Lean 80/20 format (~30 lines).
165
225
 
@@ -183,9 +243,9 @@
183
243
 
184
244
  ---
185
245
 
186
- ### BRD Structure Creation
246
+ ### STEP 5: BRD Structure Creation (ONLY if user chose "Create BRD")
187
247
 
188
- **Create project-wide** (`brd/`):
248
+ **Create project-wide** (`plans/brd/`):
189
249
  - README.md (all features index)
190
250
  - context.md (stakeholders, users, constraints)
191
251
  - references.md (industry, compliance, competitor overview)