gsd-opencode 1.3.31

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.
Files changed (68) hide show
  1. package/bin/install.js +222 -0
  2. package/command/gsd/add-phase.md +207 -0
  3. package/command/gsd/complete-milestone.md +105 -0
  4. package/command/gsd/consider-issues.md +201 -0
  5. package/command/gsd/create-roadmap.md +115 -0
  6. package/command/gsd/discuss-milestone.md +47 -0
  7. package/command/gsd/discuss-phase.md +60 -0
  8. package/command/gsd/execute-plan.md +128 -0
  9. package/command/gsd/help.md +315 -0
  10. package/command/gsd/insert-phase.md +227 -0
  11. package/command/gsd/list-phase-assumptions.md +50 -0
  12. package/command/gsd/map-codebase.md +84 -0
  13. package/command/gsd/new-milestone.md +59 -0
  14. package/command/gsd/new-project.md +316 -0
  15. package/command/gsd/pause-work.md +122 -0
  16. package/command/gsd/plan-fix.md +205 -0
  17. package/command/gsd/plan-phase.md +67 -0
  18. package/command/gsd/progress.md +316 -0
  19. package/command/gsd/remove-phase.md +338 -0
  20. package/command/gsd/research-phase.md +91 -0
  21. package/command/gsd/resume-work.md +40 -0
  22. package/command/gsd/verify-work.md +71 -0
  23. package/get-shit-done/references/checkpoints.md +287 -0
  24. package/get-shit-done/references/continuation-format.md +255 -0
  25. package/get-shit-done/references/git-integration.md +254 -0
  26. package/get-shit-done/references/plan-format.md +428 -0
  27. package/get-shit-done/references/principles.md +157 -0
  28. package/get-shit-done/references/questioning.md +162 -0
  29. package/get-shit-done/references/research-pitfalls.md +215 -0
  30. package/get-shit-done/references/scope-estimation.md +172 -0
  31. package/get-shit-done/references/tdd.md +263 -0
  32. package/get-shit-done/templates/codebase/architecture.md +255 -0
  33. package/get-shit-done/templates/codebase/concerns.md +310 -0
  34. package/get-shit-done/templates/codebase/conventions.md +307 -0
  35. package/get-shit-done/templates/codebase/integrations.md +280 -0
  36. package/get-shit-done/templates/codebase/stack.md +186 -0
  37. package/get-shit-done/templates/codebase/structure.md +285 -0
  38. package/get-shit-done/templates/codebase/testing.md +480 -0
  39. package/get-shit-done/templates/config.json +18 -0
  40. package/get-shit-done/templates/context.md +161 -0
  41. package/get-shit-done/templates/continue-here.md +78 -0
  42. package/get-shit-done/templates/discovery.md +146 -0
  43. package/get-shit-done/templates/issues.md +32 -0
  44. package/get-shit-done/templates/milestone-archive.md +123 -0
  45. package/get-shit-done/templates/milestone-context.md +93 -0
  46. package/get-shit-done/templates/milestone.md +115 -0
  47. package/get-shit-done/templates/phase-prompt.md +303 -0
  48. package/get-shit-done/templates/project.md +184 -0
  49. package/get-shit-done/templates/research.md +529 -0
  50. package/get-shit-done/templates/roadmap.md +196 -0
  51. package/get-shit-done/templates/state.md +210 -0
  52. package/get-shit-done/templates/summary.md +273 -0
  53. package/get-shit-done/templates/uat-issues.md +143 -0
  54. package/get-shit-done/workflows/complete-milestone.md +643 -0
  55. package/get-shit-done/workflows/create-milestone.md +416 -0
  56. package/get-shit-done/workflows/create-roadmap.md +481 -0
  57. package/get-shit-done/workflows/discovery-phase.md +293 -0
  58. package/get-shit-done/workflows/discuss-milestone.md +236 -0
  59. package/get-shit-done/workflows/discuss-phase.md +247 -0
  60. package/get-shit-done/workflows/execute-phase.md +1625 -0
  61. package/get-shit-done/workflows/list-phase-assumptions.md +178 -0
  62. package/get-shit-done/workflows/map-codebase.md +434 -0
  63. package/get-shit-done/workflows/plan-phase.md +488 -0
  64. package/get-shit-done/workflows/research-phase.md +436 -0
  65. package/get-shit-done/workflows/resume-project.md +287 -0
  66. package/get-shit-done/workflows/transition.md +580 -0
  67. package/get-shit-done/workflows/verify-work.md +202 -0
  68. package/package.json +38 -0
@@ -0,0 +1,293 @@
1
+ <purpose>
2
+ Execute discovery at the appropriate depth level.
3
+ Produces DISCOVERY.md (for Level 2-3) that informs PLAN.md creation.
4
+
5
+ Called from plan-phase.md's mandatory_discovery step with a depth parameter.
6
+
7
+ NOTE: For comprehensive ecosystem research ("how do experts build this"), use /gsd:research-phase instead, which produces RESEARCH.md.
8
+ </purpose>
9
+
10
+ <depth_levels>
11
+ **This workflow supports three depth levels:**
12
+
13
+ | Level | Name | Time | Output | When |
14
+ | ----- | ------------ | --------- | -------------------------------------------- | ----------------------------------------- |
15
+ | 1 | Quick Verify | 2-5 min | No file, proceed with verified knowledge | Single library, confirming current syntax |
16
+ | 2 | Standard | 15-30 min | DISCOVERY.md | Choosing between options, new integration |
17
+ | 3 | Deep Dive | 1+ hour | Detailed DISCOVERY.md with validation gates | Architectural decisions, novel problems |
18
+
19
+ **Depth is determined by plan-phase.md before routing here.**
20
+ </depth_levels>
21
+
22
+ <source_hierarchy>
23
+ **MANDATORY: Context7 BEFORE WebSearch**
24
+
25
+ Claude's training data is 6-18 months stale. Always verify.
26
+
27
+ 1. **Context7 MCP FIRST** - Current docs, no hallucination
28
+ 2. **Official docs** - When Context7 lacks coverage
29
+ 3. **WebSearch LAST** - For comparisons and trends only
30
+
31
+ See ~/.config/opencode/get-shit-done/templates/discovery.md `<discovery_protocol>` for full protocol.
32
+ </source_hierarchy>
33
+
34
+ <process>
35
+
36
+ <step name="determine_depth">
37
+ Check the depth parameter passed from plan-phase.md:
38
+ - `depth=verify` → Level 1 (Quick Verification)
39
+ - `depth=standard` → Level 2 (Standard Discovery)
40
+ - `depth=deep` → Level 3 (Deep Dive)
41
+
42
+ Route to appropriate level workflow below.
43
+ </step>
44
+
45
+ <step name="level_1_quick_verify">
46
+ **Level 1: Quick Verification (2-5 minutes)**
47
+
48
+ For: Single known library, confirming syntax/version still correct.
49
+
50
+ **Process:**
51
+
52
+ 1. Resolve library in Context7:
53
+
54
+ ```
55
+ mcp__context7__resolve-library-id with libraryName: "[library]"
56
+ ```
57
+
58
+ 2. Fetch relevant docs:
59
+
60
+ ```
61
+ mcp__context7__get-library-docs with:
62
+ - context7CompatibleLibraryID: [from step 1]
63
+ - topic: [specific concern]
64
+ ```
65
+
66
+ 3. Verify:
67
+
68
+ - Current version matches expectations
69
+ - API syntax unchanged
70
+ - No breaking changes in recent versions
71
+
72
+ 4. **If verified:** Return to plan-phase.md with confirmation. No DISCOVERY.md needed.
73
+
74
+ 5. **If concerns found:** Escalate to Level 2.
75
+
76
+ **Output:** Verbal confirmation to proceed, or escalation to Level 2.
77
+ </step>
78
+
79
+ <step name="level_2_standard">
80
+ **Level 2: Standard Discovery (15-30 minutes)**
81
+
82
+ For: Choosing between options, new external integration.
83
+
84
+ **Process:**
85
+
86
+ 1. **Identify what to discover:**
87
+
88
+ - What options exist?
89
+ - What are the key comparison criteria?
90
+ - What's our specific use case?
91
+
92
+ 2. **Context7 for each option:**
93
+
94
+ ```
95
+ For each library/framework:
96
+ - mcp__context7__resolve-library-id
97
+ - mcp__context7__get-library-docs (mode: "code" for API, "info" for concepts)
98
+ ```
99
+
100
+ 3. **Official docs** for anything Context7 lacks.
101
+
102
+ 4. **WebSearch** for comparisons:
103
+
104
+ - "[option A] vs [option B] {current_year}"
105
+ - "[option] known issues"
106
+ - "[option] with [our stack]"
107
+
108
+ 5. **Cross-verify:** Any WebSearch finding → confirm with Context7/official docs.
109
+
110
+ 6. **Quality check:** Before finalizing findings, consult ~/.config/opencode/get-shit-done/references/research-pitfalls.md to avoid common research gaps.
111
+
112
+ 7. **Create DISCOVERY.md** using ~/.config/opencode/get-shit-done/templates/discovery.md structure:
113
+
114
+ - Summary with recommendation
115
+ - Key findings per option
116
+ - Code examples from Context7
117
+ - Confidence level (should be MEDIUM-HIGH for Level 2)
118
+
119
+ 8. Return to plan-phase.md.
120
+
121
+ **Output:** `.planning/phases/XX-name/DISCOVERY.md`
122
+ </step>
123
+
124
+ <step name="level_3_deep_dive">
125
+ **Level 3: Deep Dive (1+ hour)**
126
+
127
+ For: Architectural decisions, novel problems, high-risk choices.
128
+
129
+ **Process:**
130
+
131
+ 1. **Scope the discovery** using ~/.config/opencode/get-shit-done/templates/discovery.md:
132
+
133
+ - Define clear scope
134
+ - Define include/exclude boundaries
135
+ - List specific questions to answer
136
+
137
+ 2. **Exhaustive Context7 research:**
138
+
139
+ - All relevant libraries
140
+ - Related patterns and concepts
141
+ - Multiple topics per library if needed
142
+
143
+ 3. **Official documentation deep read:**
144
+
145
+ - Architecture guides
146
+ - Best practices sections
147
+ - Migration/upgrade guides
148
+ - Known limitations
149
+
150
+ 4. **WebSearch for ecosystem context:**
151
+
152
+ - How others solved similar problems
153
+ - Production experiences
154
+ - Gotchas and anti-patterns
155
+ - Recent changes/announcements
156
+
157
+ 5. **Cross-verify ALL findings:**
158
+
159
+ - Every WebSearch claim → verify with authoritative source
160
+ - Mark what's verified vs assumed
161
+ - Flag contradictions
162
+
163
+ 6. **Quality check:** Before finalizing findings, consult ~/.config/opencode/get-shit-done/references/research-pitfalls.md to ensure comprehensive coverage and avoid common research gaps.
164
+
165
+ 7. **Create comprehensive DISCOVERY.md:**
166
+
167
+ - Full structure from ~/.config/opencode/get-shit-done/templates/discovery.md
168
+ - Quality report with source attribution
169
+ - Confidence by finding
170
+ - If LOW confidence on any critical finding → add validation checkpoints
171
+
172
+ 8. **Confidence gate:** If overall confidence is LOW, present options before proceeding.
173
+
174
+ 9. Return to plan-phase.md.
175
+
176
+ **Output:** `.planning/phases/XX-name/DISCOVERY.md` (comprehensive)
177
+ </step>
178
+
179
+ <step name="identify_unknowns">
180
+ **For Level 2-3:** Define what we need to learn.
181
+
182
+ Ask: What do we need to learn before we can plan this phase?
183
+
184
+ - Technology choices?
185
+ - Best practices?
186
+ - API patterns?
187
+ - Architecture approach?
188
+ </step>
189
+
190
+ <step name="create_discovery_scope">
191
+ Use ~/.config/opencode/get-shit-done/templates/discovery.md.
192
+
193
+ Include:
194
+
195
+ - Clear discovery objective
196
+ - Scoped include/exclude lists
197
+ - Source preferences (official docs, Context7, current year)
198
+ - Output structure for DISCOVERY.md
199
+ </step>
200
+
201
+ <step name="execute_discovery">
202
+ Run the discovery:
203
+ - Use web search for current info
204
+ - Use Context7 MCP for library docs
205
+ - Prefer current year sources
206
+ - Structure findings per template
207
+ </step>
208
+
209
+ <step name="create_discovery_output">
210
+ Write `.planning/phases/XX-name/DISCOVERY.md`:
211
+ - Summary with recommendation
212
+ - Key findings with sources
213
+ - Code examples if applicable
214
+ - Metadata (confidence, dependencies, open questions, assumptions)
215
+ </step>
216
+
217
+ <step name="confidence_gate">
218
+ After creating DISCOVERY.md, check confidence level.
219
+
220
+ If confidence is LOW:
221
+ Use AskUserQuestion:
222
+
223
+ - header: "Low Confidence"
224
+ - question: "Discovery confidence is LOW: [reason]. How would you like to proceed?"
225
+ - options:
226
+ - "Dig deeper" - Do more research before planning
227
+ - "Proceed anyway" - Accept uncertainty, plan with caveats
228
+ - "Pause" - I need to think about this
229
+
230
+ If confidence is MEDIUM:
231
+ Inline: "Discovery complete (medium confidence). [brief reason]. Proceed to planning?"
232
+
233
+ If confidence is HIGH:
234
+ Proceed directly, just note: "Discovery complete (high confidence)."
235
+ </step>
236
+
237
+ <step name="open_questions_gate">
238
+ If DISCOVERY.md has open_questions:
239
+
240
+ Present them inline:
241
+ "Open questions from discovery:
242
+
243
+ - [Question 1]
244
+ - [Question 2]
245
+
246
+ These may affect implementation. Acknowledge and proceed? (yes / address first)"
247
+
248
+ If "address first": Gather user input on questions, update discovery.
249
+ </step>
250
+
251
+ <step name="offer_next">
252
+ ```
253
+ Discovery complete: .planning/phases/XX-name/DISCOVERY.md
254
+ Recommendation: [one-liner]
255
+ Confidence: [level]
256
+
257
+ What's next?
258
+
259
+ 1. Discuss phase context (/gsd:discuss-phase [current-phase])
260
+ 2. Create phase plan (/gsd:plan-phase [current-phase])
261
+ 3. Refine discovery (dig deeper)
262
+ 4. Review discovery
263
+
264
+ ```
265
+
266
+ NOTE: DISCOVERY.md is NOT committed separately. It will be committed with phase completion.
267
+ </step>
268
+
269
+ </process>
270
+
271
+ <success_criteria>
272
+ **Level 1 (Quick Verify):**
273
+ - Context7 consulted for library/topic
274
+ - Current state verified or concerns escalated
275
+ - Verbal confirmation to proceed (no files)
276
+
277
+ **Level 2 (Standard):**
278
+ - Context7 consulted for all options
279
+ - WebSearch findings cross-verified
280
+ - DISCOVERY.md created with recommendation
281
+ - Confidence level MEDIUM or higher
282
+ - Ready to inform PLAN.md creation
283
+
284
+ **Level 3 (Deep Dive):**
285
+ - Discovery scope defined
286
+ - Context7 exhaustively consulted
287
+ - All WebSearch findings verified against authoritative sources
288
+ - DISCOVERY.md created with comprehensive analysis
289
+ - Quality report with source attribution
290
+ - If LOW confidence findings → validation checkpoints defined
291
+ - Confidence gate passed
292
+ - Ready to inform PLAN.md creation
293
+ </success_criteria>
@@ -0,0 +1,236 @@
1
+ <purpose>
2
+ Help the user figure out what they want to build in the next milestone through collaborative thinking.
3
+
4
+ You're a thinking partner helping them crystallize their vision for what's next. Features first — everything else (scope, phases) derives from what they want to build.
5
+ </purpose>
6
+
7
+ <process>
8
+
9
+ <step name="check_state" priority="first">
10
+ Load project state:
11
+
12
+ ```bash
13
+ cat .planning/STATE.md
14
+ cat .planning/ROADMAP.md
15
+ ```
16
+
17
+ **If no active milestone (expected state after completing previous):**
18
+ Continue to milestone_context.
19
+
20
+ **If active milestone exists:**
21
+
22
+ ```
23
+ Current milestone in progress: v[X.Y] [Name]
24
+ Phases [N]-[M], [P]% complete
25
+
26
+ Did you want to:
27
+ 1. Complete current milestone first (/gsd:complete-milestone)
28
+ 2. Add phases to current milestone (/gsd:add-phase)
29
+ 3. Continue anyway - discuss next milestone scope
30
+
31
+ ```
32
+
33
+ Wait for user response. If "Continue anyway", proceed to milestone_context.
34
+ </step>
35
+
36
+ <step name="milestone_context">
37
+ Present context from previous milestone:
38
+
39
+ ```
40
+ Last completed: v[X.Y] [Name] (shipped [DATE])
41
+ Key accomplishments:
42
+ - [From MILESTONES.md or STATE.md]
43
+
44
+ Total phases delivered: [N]
45
+ Next phase number: [N+1]
46
+ ```
47
+
48
+ Continue to intake_gate.
49
+ </step>
50
+
51
+ <step name="intake_gate">
52
+ **CRITICAL: ALL questions use AskUserQuestion. Never ask inline text questions.**
53
+
54
+ The primary question is: **What do you want to build/add/fix?**
55
+
56
+ Everything else (scope, priority, constraints) is secondary and derived from features.
57
+
58
+ Check for inputs:
59
+ - Deferred issues from STATE.md (potential features)
60
+ - Known gaps or pain points from usage
61
+ - User's ideas for what's next
62
+
63
+ **1. Open:**
64
+
65
+ Use AskUserQuestion:
66
+ - header: "Next"
67
+ - question: "What do you want to add, improve, or fix in this milestone?"
68
+ - options: [Deferred issues from STATE.md if any] + ["New features", "Improvements to existing", "Bug fixes", "Let me describe"]
69
+
70
+ **2. Explore features:**
71
+
72
+ Based on their response, use AskUserQuestion:
73
+
74
+ If they named specific features:
75
+ - header: "Feature Details"
76
+ - question: "Tell me more about [feature] - what should it do?"
77
+ - options: [Contextual options based on feature type + "Let me describe it"]
78
+
79
+ If they described a general direction:
80
+ - header: "Breaking It Down"
81
+ - question: "That could involve [A], [B], [C] - which matter most?"
82
+ - options: [Specific sub-features + "All of them" + "Something else"]
83
+
84
+ If they're not sure:
85
+ - header: "Starting Points"
86
+ - question: "What's been frustrating or missing?"
87
+ - options: [Deferred issues from STATE.md + pain point categories + "Let me think about it"]
88
+
89
+ **3. Prioritize:**
90
+
91
+ Use AskUserQuestion:
92
+ - header: "Priority"
93
+ - question: "Which of these matters most?"
94
+ - options: [Features they mentioned + "All equally important" + "Let me prioritize"]
95
+
96
+ After gathering features, synthesize:
97
+
98
+ ```
99
+ Based on what you described:
100
+
101
+ **Features:**
102
+ - [Feature 1]: [brief description]
103
+ - [Feature 2]: [brief description]
104
+ - [Feature 3]: [brief description]
105
+
106
+ **Estimated scope:** [N] phases
107
+ **Theme suggestion:** v[X.Y] [Name]
108
+ ```
109
+
110
+ **4. Decision gate:**
111
+
112
+ Use AskUserQuestion:
113
+ - header: "Ready?"
114
+ - question: "Ready to create the milestone, or explore more?"
115
+ - options (ALL THREE REQUIRED):
116
+ - "Create milestone" - Proceed to /gsd:new-milestone
117
+ - "Ask more questions" - Help me think through this more
118
+ - "Let me add context" - I have more to share
119
+
120
+ If "Ask more questions" → return to step 2 with new probes.
121
+ If "Let me add context" → receive input → return to step 2.
122
+ Loop until "Create milestone" selected.
123
+ </step>
124
+
125
+ <step name="write_context">
126
+ Write milestone context to file for handoff.
127
+
128
+ **File:** `.planning/MILESTONE-CONTEXT.md`
129
+
130
+ Use template from ~/.config/opencode/get-shit-done/templates/milestone-context.md
131
+
132
+ Populate with:
133
+ - Features identified during discussion
134
+ - Suggested milestone name and theme
135
+ - Estimated phase count
136
+ - How features map to phases
137
+ - Any constraints or scope boundaries mentioned
138
+
139
+ ```bash
140
+ # Write the context file
141
+ cat > .planning/MILESTONE-CONTEXT.md << 'EOF'
142
+ # Milestone Context
143
+
144
+ **Generated:** [today's date]
145
+ **Status:** Ready for /gsd:new-milestone
146
+
147
+ <features>
148
+ ## Features to Build
149
+
150
+ - **[Feature 1]**: [description]
151
+ - **[Feature 2]**: [description]
152
+ - **[Feature 3]**: [description]
153
+
154
+ </features>
155
+
156
+ <scope>
157
+ ## Scope
158
+
159
+ **Suggested name:** v[X.Y] [Theme Name]
160
+ **Estimated phases:** [N]
161
+ **Focus:** [One sentence theme/focus]
162
+
163
+ </scope>
164
+
165
+ <phase_mapping>
166
+ ## Phase Mapping
167
+
168
+ - Phase [N]: [Feature/goal]
169
+ - Phase [N+1]: [Feature/goal]
170
+ - Phase [N+2]: [Feature/goal]
171
+
172
+ </phase_mapping>
173
+
174
+ <constraints>
175
+ ## Constraints
176
+
177
+ - [Any constraints mentioned]
178
+
179
+ </constraints>
180
+
181
+ <notes>
182
+ ## Additional Context
183
+
184
+ [Anything else from discussion]
185
+
186
+ </notes>
187
+
188
+ ---
189
+
190
+ *This file is temporary. It will be deleted after /gsd:new-milestone creates the milestone.*
191
+ EOF
192
+ ```
193
+ </step>
194
+
195
+ <step name="handoff">
196
+ Present summary and hand off to create-milestone:
197
+
198
+ ```
199
+ Milestone scope defined:
200
+
201
+ **Features:**
202
+ - [Feature 1]: [description]
203
+ - [Feature 2]: [description]
204
+ - [Feature 3]: [description]
205
+
206
+ **Suggested milestone:** v[X.Y] [Theme Name]
207
+ **Estimated phases:** [N]
208
+
209
+ Context saved to `.planning/MILESTONE-CONTEXT.md`
210
+
211
+ ---
212
+
213
+ ## ▶ Next Up
214
+
215
+ **Create Milestone v[X.Y]** — [Theme Name]
216
+
217
+ `/gsd:new-milestone`
218
+
219
+ <sub>`/clear` first → fresh context window</sub>
220
+
221
+ ---
222
+ ```
223
+ </step>
224
+
225
+ </process>
226
+
227
+ <success_criteria>
228
+
229
+ - Project state loaded (STATE.md, ROADMAP.md)
230
+ - Previous milestone context presented
231
+ - **Features identified** - What to build/add/fix (the substance)
232
+ - Features explored with clarifying questions
233
+ - Scope synthesized from features (not asked abstractly)
234
+ - **MILESTONE-CONTEXT.md created** with features, scope, and phase mapping
235
+ - Context handed off to /gsd:new-milestone
236
+ </success_criteria>