@leeovery/claude-technical-workflows 2.0.10 → 2.0.12

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.
@@ -1,36 +1,321 @@
1
1
  ---
2
- description: Start a technical discussion using the technical-discussion skill. Gathers information about the topic and creates discussion documentation in docs/workflow/discussion/
2
+ description: Start a technical discussion. Discovers research and existing discussions, offers multiple entry paths, and invokes the technical-discussion skill.
3
3
  ---
4
4
 
5
5
  Invoke the **technical-discussion** skill for this conversation.
6
6
 
7
- Before beginning, ask the user these questions to properly set up the discussion:
7
+ ## Instructions
8
8
 
9
- ## Essential Information
9
+ Follow these steps EXACTLY as written. Do not skip steps or combine them. Present output using the EXACT format shown in examples - do not simplify or alter the formatting.
10
10
 
11
- 1. **Discussion Topic**: What are we discussing? (This will be used as the filename in `docs/workflow/discussion/{topic}.md`)
11
+ Before beginning, discover existing work and determine the best entry path.
12
12
 
13
- 2. **Context & Background**:
14
- - What sparked this discussion?
15
- - What problem are we trying to solve?
16
- - Why are we discussing this now?
13
+ ## Important
17
14
 
18
- 3. **Foundational Knowledge**:
19
- - Is there any background information I need to understand before we begin?
20
- - Are there specific concepts, technologies, or architectures I should know about?
21
- - Any constraints, requirements, or limitations I should be aware of?
15
+ Use simple, individual commands. Never combine multiple operations into bash loops or one-liners. Execute commands one at a time.
22
16
 
23
- 4. **Codebase Review**:
24
- - Are there specific files in this repository I should read first?
25
- - Should I explore any particular directories or components?
26
- - Is there existing code related to this discussion?
17
+ ## Step 1: Discover Existing Work
27
18
 
28
- ## After Gathering Information
19
+ Scan the codebase for research and discussions:
29
20
 
30
- Once I have this information:
21
+ 1. **Find research**: Look in `docs/workflow/research/`
22
+ - Run `ls docs/workflow/research/` to list research files
23
+ - Note which files exist (may include `exploration.md` and semantic files like `market-landscape.md`)
24
+
25
+ 2. **Find discussions**: Look in `docs/workflow/discussion/`
26
+ - Run `ls docs/workflow/discussion/` to list discussion files
27
+ - Each file is named `{topic}.md`
28
+
29
+ 3. **Check discussion status**: For each discussion file
30
+ - Run `head -10 docs/workflow/discussion/{topic}.md` to extract the `Status:` field
31
+ - Status values: `Exploring`, `Deciding`, or `Concluded`
32
+ - Do NOT use bash loops - run separate commands for each file
33
+
34
+ 4. **Check for cached analysis** (if research files exist):
35
+ - Check if `docs/workflow/.cache/research-analysis.md` exists
36
+ - If it exists, read it to get the stored checksum from the frontmatter
37
+
38
+ 5. **Compute current research checksum** (if research files exist):
39
+ - Run: `cat $(ls docs/workflow/research/*.md | sort) 2>/dev/null | md5sum | cut -d' ' -f1`
40
+ - This creates a combined checksum of all research files (sorted alphabetically)
41
+ - Store this value to compare with the cached checksum
42
+
43
+ ## Step 2: Present Workflow State and Options
44
+
45
+ Present the workflow state and available options based on what was discovered.
46
+
47
+ **Format:**
48
+ ```
49
+ 📂 Workflow state:
50
+ 📚 Research: {count} files found / None found
51
+ 💬 Discussions: {count} existing / None yet
52
+ ```
53
+
54
+ Then present the appropriate options:
55
+
56
+ **If research AND discussions exist:**
57
+ ```
58
+ How would you like to proceed?
59
+
60
+ 1. **From research** - Analyze research and suggest undiscussed topics
61
+ 2. **Continue discussion** - Resume an existing discussion
62
+ 3. **Fresh topic** - Start a new discussion
63
+ ```
64
+
65
+ **If ONLY research exists:**
66
+ ```
67
+ How would you like to proceed?
68
+
69
+ 1. **From research** - Analyze research and suggest topics to discuss
70
+ 2. **Fresh topic** - Start a new discussion
71
+ ```
72
+
73
+ **If ONLY discussions exist:**
74
+ ```
75
+ How would you like to proceed?
76
+
77
+ 1. **Continue discussion** - Resume an existing discussion
78
+ 2. **Fresh topic** - Start a new discussion
79
+ ```
80
+
81
+ **If NOTHING exists:**
82
+ ```
83
+ Starting fresh - no prior research or discussions found.
84
+
85
+ What topic would you like to discuss?
86
+ ```
87
+ Then skip to Step 5 (Fresh topic path).
88
+
89
+ Wait for the user to choose before proceeding.
90
+
91
+ ## Step 3A: "From research" Path
92
+
93
+ This step uses caching to avoid re-analyzing unchanged research documents.
94
+
95
+ ### Step 3A.1: Check Cache Validity
96
+
97
+ Compare the current research checksum (computed in Step 1.5) with the cached checksum:
98
+
99
+ **If cache exists AND checksums match:**
100
+ ```
101
+ 📋 Using cached analysis
102
+
103
+ Research documents unchanged since last analysis ({date from cache}).
104
+ Loading {count} previously identified topics...
105
+
106
+ 💡 To force a fresh analysis, enter: refresh
107
+ ```
108
+
109
+ Then load the topics from the cache file and skip to Step 3A.3 (Cross-reference).
110
+
111
+ **If cache missing OR checksums differ:**
112
+ ```
113
+ 🔍 Analyzing research documents...
114
+ ```
115
+
116
+ Proceed to Step 3A.2 (Full Analysis).
117
+
118
+ ### Step 3A.2: Full Analysis (when cache invalid)
119
+
120
+ Read each research file and analyze the content to extract key themes and potential discussion topics. For each theme:
121
+ - Note the source file and relevant line numbers
122
+ - Summarize what the theme is about in 1-2 sentences
123
+ - Identify key questions or decisions that need discussion
124
+
125
+ **Be thorough**: This analysis will be cached, so take time to identify ALL potential topics including:
126
+ - Major architectural decisions
127
+ - Technical trade-offs mentioned
128
+ - Open questions or concerns raised
129
+ - Implementation approaches discussed
130
+ - Integration points with external systems
131
+ - Security or performance considerations
132
+ - Edge cases or error handling mentioned
133
+
134
+ **Save to cache:**
135
+ After analysis, create/update `docs/workflow/.cache/research-analysis.md`:
136
+
137
+ ```markdown
138
+ ---
139
+ checksum: {computed_checksum}
140
+ generated: {ISO date}
141
+ research_files:
142
+ - {filename1}.md
143
+ - {filename2}.md
144
+ ---
145
+
146
+ # Research Analysis Cache
147
+
148
+ ## Topics
149
+
150
+ ### {Theme name}
151
+ - **Source**: {filename}.md (lines {start}-{end})
152
+ - **Summary**: {1-2 sentence summary}
153
+ - **Key questions**: {what needs deciding}
154
+
155
+ ### {Another theme}
156
+ - **Source**: {filename}.md (lines {start}-{end})
157
+ - **Summary**: {1-2 sentence summary}
158
+ - **Key questions**: {what needs deciding}
159
+
160
+ [... more topics ...]
161
+ ```
162
+
163
+ Ensure the `.cache` directory exists: `mkdir -p docs/workflow/.cache`
164
+
165
+ ### Step 3A.3: Cross-reference with Discussions
166
+
167
+ **Always performed** (whether using cache or fresh analysis):
168
+
169
+ For each identified topic, check if a corresponding discussion already exists in `docs/workflow/discussion/`.
170
+
171
+ ### Step 3A.4: Present Findings
172
+
173
+ **If using cached analysis:**
174
+ ```
175
+ 📋 Cached analysis (research unchanged since {date})
176
+
177
+ 💡 Topics identified:
178
+
179
+ ✨ {Theme name}
180
+ Source: {filename}.md (lines {start}-{end})
181
+ "{Brief summary}"
182
+
183
+ ✅ {Already discussed theme} → discussed in {topic}.md
184
+ Source: {filename}.md (lines {start}-{end})
185
+ "{Brief summary}"
186
+
187
+ Which topic would you like to discuss? (Or enter 'refresh' for fresh analysis)
188
+ ```
189
+
190
+ **If fresh analysis:**
191
+ ```
192
+ 🔍 Analysis complete (cached for future sessions)
193
+
194
+ 💡 Topics identified:
195
+
196
+ ✨ {Theme name}
197
+ Source: {filename}.md (lines {start}-{end})
198
+ "{Brief summary}"
199
+
200
+ ✅ {Already discussed theme} → discussed in {topic}.md
201
+ Source: {filename}.md (lines {start}-{end})
202
+ "{Brief summary}"
203
+
204
+ Which topic would you like to discuss? (Or describe something else)
205
+ ```
206
+
207
+ **Key:**
208
+ - ✨ = Undiscussed topic (potential new discussion)
209
+ - ✅ = Already has a corresponding discussion
210
+
211
+ ### Step 3A.5: Handle "refresh" Request
212
+
213
+ If user enters `refresh`:
214
+ - Delete the cache file: `rm docs/workflow/.cache/research-analysis.md`
215
+ - Return to Step 3A.2 (Full Analysis)
216
+ - Inform user: "Refreshing analysis..."
217
+
218
+ **Important:** Keep track of the source file and line numbers for the chosen topic - this will be passed to the skill.
219
+
220
+ Wait for the user to choose before proceeding to Step 4.
221
+
222
+ ## Step 3B: "Continue discussion" Path
223
+
224
+ List existing discussions with their status:
225
+
226
+ ```
227
+ 💬 Existing discussions:
228
+
229
+ ⚡ {topic}.md — {Status}
230
+ "{Brief description from context section}"
231
+
232
+ ⚡ {topic}.md — {Status}
233
+ "{Brief description}"
234
+
235
+ ✅ {topic}.md — Concluded
236
+ "{Brief description}"
237
+
238
+ Which discussion would you like to continue?
239
+ ```
240
+
241
+ **Key:**
242
+ - ⚡ = In progress (Exploring or Deciding)
243
+ - ✅ = Concluded (can still be continued/reopened)
244
+
245
+ Wait for the user to choose, then proceed to Step 4.
246
+
247
+ ## Step 3C: "Fresh topic" Path
248
+
249
+ Proceed directly to Step 4.
250
+
251
+ ## Step 4: Gather Context
252
+
253
+ Gather context based on the chosen path.
254
+
255
+ **If starting new discussion (from research or fresh):**
256
+
257
+ ```
258
+ ## New discussion: {topic}
259
+
260
+ Before we begin:
261
+
262
+ 1. What's the core problem or decision we need to work through?
263
+
264
+ 2. Any constraints or context I should know about?
265
+
266
+ 3. Are there specific files in the codebase I should review first?
267
+ ```
268
+
269
+ Wait for responses before proceeding.
270
+
271
+ **If continuing existing discussion:**
272
+
273
+ Read the existing discussion document first, then ask:
274
+
275
+ ```
276
+ ## Continuing: {topic}
277
+
278
+ I've read the existing discussion.
279
+
280
+ What would you like to focus on in this session?
281
+ ```
282
+
283
+ Wait for response before proceeding.
284
+
285
+ ## Step 5: Invoke Discussion Skill
286
+
287
+ Begin the discussion session with appropriate context based on the path taken.
288
+
289
+ **If from research:**
290
+ ```
291
+ Discussion session for: {topic}
292
+ Output: docs/workflow/discussion/{topic}.md
293
+
294
+ ## Research Reference
295
+ Source: docs/workflow/research/{filename}.md (lines {start}-{end})
296
+ Summary: {the 1-2 sentence summary from Step 3A}
297
+
298
+ Begin discussion using the technical-discussion skill.
299
+ ```
300
+
301
+ **If continuing or fresh:**
302
+ ```
303
+ Discussion session for: {topic}
304
+ Source: {existing discussion | fresh}
305
+ Output: docs/workflow/discussion/{topic}.md
306
+
307
+ Begin discussion using the technical-discussion skill.
308
+ ```
309
+
310
+ **Setup:**
31
311
  - Ensure discussion directory exists: `docs/workflow/discussion/`
32
- - Start documenting the discussion following the technical-discussion skill structure
33
- - Create file: `docs/workflow/discussion/{topic}.md`
312
+ - If new: Create file using the template structure
313
+ - If continuing: Work with existing file
34
314
  - Commit frequently at natural discussion breaks
35
315
 
36
- Ask these questions clearly and wait for responses before proceeding with the discussion.
316
+ ## Notes
317
+
318
+ - Ask questions clearly and wait for responses before proceeding
319
+ - Discussion captures WHAT and WHY - don't jump to specifications or implementation
320
+ - The goal is to work through edge cases, debates, and decisions before planning
321
+ - Commit the discussion document frequently during the session
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@leeovery/claude-technical-workflows",
3
- "version": "2.0.10",
3
+ "version": "2.0.12",
4
4
  "description": "Technical workflow skills & commands for Claude Code",
5
5
  "license": "MIT",
6
6
  "author": "Lee Overy <me@leeovery.com>",