specdacular 0.7.0 → 0.7.1

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 (36) hide show
  1. package/README.md +39 -86
  2. package/commands/specd/feature/{next.md → continue.md} +6 -6
  3. package/commands/specd/feature/new.md +2 -2
  4. package/commands/specd/feature/toolbox.md +49 -0
  5. package/commands/specd/help.md +1 -146
  6. package/commands/specd/update.md +4 -6
  7. package/package.json +1 -1
  8. package/specdacular/HELP.md +84 -0
  9. package/specdacular/references/commit-code.md +34 -0
  10. package/specdacular/references/commit-docs.md +35 -0
  11. package/specdacular/references/select-feature.md +57 -0
  12. package/specdacular/references/select-phase.md +30 -0
  13. package/specdacular/templates/features/STATE.md +11 -1
  14. package/specdacular/workflows/{next-feature.md → continue-feature.md} +50 -90
  15. package/specdacular/workflows/discuss-feature.md +4 -27
  16. package/specdacular/workflows/execute-plan.md +75 -47
  17. package/specdacular/workflows/insert-phase.md +50 -26
  18. package/specdacular/workflows/map-codebase.md +4 -30
  19. package/specdacular/workflows/new-feature.md +14 -38
  20. package/specdacular/workflows/plan-feature.md +6 -29
  21. package/specdacular/workflows/plan-phase.md +4 -29
  22. package/specdacular/workflows/prepare-phase.md +4 -35
  23. package/specdacular/workflows/research-phase.md +4 -30
  24. package/specdacular/workflows/review-feature.md +316 -0
  25. package/specdacular/workflows/review-phase.md +4 -30
  26. package/specdacular/workflows/toolbox.md +115 -0
  27. package/commands/specd/feature/discuss.md +0 -66
  28. package/commands/specd/feature/plan.md +0 -68
  29. package/commands/specd/feature/research.md +0 -459
  30. package/commands/specd/phase/execute.md +0 -68
  31. package/commands/specd/phase/insert.md +0 -62
  32. package/commands/specd/phase/plan.md +0 -73
  33. package/commands/specd/phase/prepare.md +0 -75
  34. package/commands/specd/phase/renumber.md +0 -66
  35. package/commands/specd/phase/research.md +0 -65
  36. package/commands/specd/phase/review.md +0 -80
@@ -1,459 +0,0 @@
1
- ---
2
- name: specd:feature:research
3
- description: Research how to implement a feature - spawns parallel agents for codebase, external, and pitfalls research
4
- argument-hint: "[feature-name]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Bash
9
- - Glob
10
- - Grep
11
- - Task
12
- - AskUserQuestion
13
- ---
14
-
15
- <objective>
16
- > **Note:** Consider using `/specd:feature:next` instead — it guides you through the entire feature lifecycle automatically, including research.
17
-
18
- Research how to implement a feature before planning. Spawns parallel research agents to investigate:
19
-
20
- 1. **Codebase Integration** - How does this fit with existing code? (uses Claude Code Explore agent)
21
- 2. **External Patterns** - What libraries/patterns are standard for this?
22
- 3. **Pitfalls** - What commonly goes wrong?
23
-
24
- **Output:** `.specd/features/{name}/RESEARCH.md` with prescriptive guidance for the planner.
25
-
26
- **Why parallel agents:** Research burns context fast. Fresh context per dimension = thorough investigation. Synthesize results into single RESEARCH.md.
27
- </objective>
28
-
29
- <execution_context>
30
- @~/.claude/specdacular/workflows/research-feature.md
31
- </execution_context>
32
-
33
- <context>
34
- Feature name: $ARGUMENTS
35
-
36
- **Load feature context:**
37
- @.specd/features/{name}/FEATURE.md
38
- @.specd/features/{name}/DECISIONS.md
39
- @.specd/features/{name}/CONTEXT.md (if exists)
40
-
41
- **Load codebase context:**
42
- @.specd/codebase/ARCHITECTURE.md
43
- @.specd/codebase/CONVENTIONS.md
44
- @.specd/codebase/STACK.md
45
- </context>
46
-
47
- <process>
48
-
49
- ## Phase 1: Validate Feature
50
-
51
- ```bash
52
- # Check feature exists
53
- [ -d ".specd/features/$ARGUMENTS" ] || { echo "Feature not found. Run /specd:feature:new first."; exit 1; }
54
-
55
- # Check if RESEARCH.md exists
56
- [ -f ".specd/features/$ARGUMENTS/RESEARCH.md" ] && echo "existing"
57
- ```
58
-
59
- **If RESEARCH.md exists:**
60
- Use AskUserQuestion:
61
- - header: "Research Exists"
62
- - question: "Research already exists for this feature. What would you like to do?"
63
- - options:
64
- - "Update research" — Re-run research, incorporate new findings
65
- - "View existing" — Show current RESEARCH.md
66
- - "Continue anyway" — Skip research, use existing
67
-
68
- ## Phase 2: Load Feature Context
69
-
70
- Read and parse:
71
- - `FEATURE.md` - What needs to be built
72
- - `DECISIONS.md` - Constraints already decided
73
- - `CONTEXT.md` - Gray areas already resolved (if exists)
74
-
75
- Extract:
76
- - Technical requirements (files to create, integrations needed)
77
- - Locked decisions (don't research alternatives)
78
- - Open questions (areas to investigate)
79
-
80
- ## Phase 3: Identify Research Dimensions
81
-
82
- Based on feature requirements, determine what needs research:
83
-
84
- **Always research:**
85
- - Codebase integration patterns
86
- - Pitfalls for this type of feature
87
-
88
- **Conditionally research:**
89
- - External libraries (if new dependencies needed)
90
- - API patterns (if building APIs)
91
- - UI patterns (if building UI components)
92
- - Data modeling (if new models needed)
93
-
94
- Present research plan:
95
- ```
96
- I'll research these dimensions for {feature-name}:
97
-
98
- 1. **Codebase Integration** — How this fits with existing code
99
- - Integration points in current architecture
100
- - Patterns to follow from existing code
101
- - Files that will need modification
102
-
103
- 2. **External Patterns** — Standard approaches for {feature-type}
104
- - Libraries commonly used
105
- - Architecture patterns
106
- - Code examples
107
-
108
- 3. **Pitfalls** — What commonly goes wrong
109
- - Implementation mistakes
110
- - Integration gotchas
111
- - Performance traps
112
-
113
- Estimated: 3 parallel research agents
114
-
115
- [Start research] / [Adjust scope]
116
- ```
117
-
118
- ## Phase 4: Spawn Parallel Research Agents
119
-
120
- ### Agent 1: Codebase Integration (using Explore agent)
121
-
122
- ```
123
- Task(
124
- prompt="Research how {feature-name} should integrate with the existing codebase.
125
-
126
- <feature_context>
127
- {FEATURE.md content}
128
- </feature_context>
129
-
130
- <locked_decisions>
131
- {Relevant decisions from DECISIONS.md}
132
- </locked_decisions>
133
-
134
- <research_questions>
135
- 1. What existing files/modules will this feature need to import from?
136
- 2. What patterns do similar features in this codebase follow?
137
- 3. Where should new files be created based on project structure?
138
- 4. What types/interfaces already exist that this feature should use?
139
- 5. What utility functions or hooks can be reused?
140
- </research_questions>
141
-
142
- <output_format>
143
- Return findings as structured markdown:
144
-
145
- ## Codebase Integration Findings
146
-
147
- ### Import Dependencies
148
- - `path/to/file` — what it provides, why needed
149
-
150
- ### Patterns to Follow
151
- - Pattern name: description, example file reference
152
-
153
- ### File Locations
154
- - New files should go in: path/reason
155
- - Existing files to modify: path/what change
156
-
157
- ### Reusable Code
158
- - Types: list with paths
159
- - Utilities: list with paths
160
- - Components: list with paths
161
-
162
- ### Integration Points
163
- - Where this connects to existing code
164
- - What interfaces to implement
165
- </output_format>
166
- ",
167
- subagent_type="Explore",
168
- description="Codebase integration research"
169
- )
170
- ```
171
-
172
- ### Agent 2: External Patterns Research
173
-
174
- ```
175
- Task(
176
- prompt="First, read ~/.claude/specdacular/agents/feature-researcher.md for your role.
177
-
178
- <research_type>
179
- External patterns research for {feature-type}.
180
- </research_type>
181
-
182
- <feature_context>
183
- {FEATURE.md content}
184
- </feature_context>
185
-
186
- <codebase_stack>
187
- {From .specd/codebase/STACK.md}
188
- </codebase_stack>
189
-
190
- <locked_decisions>
191
- {Decisions that constrain library/pattern choices}
192
- </locked_decisions>
193
-
194
- <research_questions>
195
- 1. What's the standard approach for {feature-type} in {stack}?
196
- 2. What libraries are commonly used? (verify versions with Context7)
197
- 3. What architecture patterns work well?
198
- 4. What code patterns are recommended?
199
- 5. What should NOT be hand-rolled?
200
- </research_questions>
201
-
202
- <tool_strategy>
203
- 1. Context7 first for any library questions
204
- 2. Official docs via WebFetch for gaps
205
- 3. WebSearch for ecosystem patterns (include current year)
206
- 4. Verify all findings - no LOW confidence recommendations
207
- </tool_strategy>
208
-
209
- <output_format>
210
- Return findings as structured markdown with confidence levels.
211
- </output_format>
212
- ",
213
- subagent_type="general-purpose",
214
- model="sonnet",
215
- description="External patterns research"
216
- )
217
- ```
218
-
219
- ### Agent 3: Pitfalls Research
220
-
221
- ```
222
- Task(
223
- prompt="First, read ~/.claude/specdacular/agents/feature-researcher.md for your role.
224
-
225
- <research_type>
226
- Pitfalls research for {feature-type}.
227
- </research_type>
228
-
229
- <feature_context>
230
- {FEATURE.md content}
231
- </feature_context>
232
-
233
- <codebase_context>
234
- {Relevant architecture/conventions}
235
- </codebase_context>
236
-
237
- <research_questions>
238
- 1. What do developers commonly get wrong with {feature-type}?
239
- 2. What are the performance pitfalls?
240
- 3. What security issues should be avoided?
241
- 4. What integration mistakes happen?
242
- 5. What looks simple but is actually complex?
243
- </research_questions>
244
-
245
- <tool_strategy>
246
- 1. WebSearch for common mistakes (include current year)
247
- 2. Look for post-mortems, issue discussions
248
- 3. Check official docs for warnings/caveats
249
- 4. Consider edge cases specific to this stack
250
- </tool_strategy>
251
-
252
- <output_format>
253
- Return pitfalls as structured markdown:
254
-
255
- ## Pitfalls for {feature-type}
256
-
257
- ### Critical (causes failures/rewrites)
258
- - Pitfall: description
259
- - Why it happens: reason
260
- - Prevention: how to avoid
261
- - Detection: warning signs
262
-
263
- ### Moderate (causes bugs/debt)
264
- - Pitfall: description
265
- - Prevention: how to avoid
266
-
267
- ### Minor (causes friction)
268
- - Pitfall: description
269
- - Prevention: how to avoid
270
- </output_format>
271
- ",
272
- subagent_type="general-purpose",
273
- model="sonnet",
274
- description="Pitfalls research"
275
- )
276
- ```
277
-
278
- ## Phase 5: Synthesize Results
279
-
280
- After all agents complete, synthesize into single RESEARCH.md:
281
-
282
- ```markdown
283
- # Research: {feature-name}
284
-
285
- **Researched:** {date}
286
- **Feature:** {feature-type}
287
- **Confidence:** {overall level}
288
-
289
- ## Summary
290
-
291
- {2-3 paragraphs synthesizing all findings}
292
-
293
- **Key recommendation:** {one-liner actionable guidance}
294
-
295
- ---
296
-
297
- ## Codebase Integration
298
-
299
- {From Agent 1 - Explore findings}
300
-
301
- ### Import From
302
- | Module | Provides | Path |
303
- |--------|----------|------|
304
- | ... | ... | ... |
305
-
306
- ### Patterns to Follow
307
- {Patterns with file references}
308
-
309
- ### File Structure
310
- ```
311
- src/
312
- ├── {where new files go}
313
- └── {existing files to modify}
314
- ```
315
-
316
- ### Reusable Code
317
- - **Types:** {list with @paths}
318
- - **Utilities:** {list with @paths}
319
- - **Components:** {list with @paths}
320
-
321
- ---
322
-
323
- ## Implementation Approach
324
-
325
- {From Agent 2 - External patterns}
326
-
327
- ### Standard Stack
328
- | Library | Version | Purpose | Confidence |
329
- |---------|---------|---------|------------|
330
- | ... | ... | ... | HIGH/MED |
331
-
332
- ### Architecture Pattern
333
- {Recommended pattern with rationale}
334
-
335
- ### Code Patterns
336
- ```typescript
337
- // Pattern name - source: {Context7/docs URL}
338
- {code example}
339
- ```
340
-
341
- ### Don't Hand-Roll
342
- | Problem | Use Instead | Why |
343
- |---------|-------------|-----|
344
- | ... | ... | ... |
345
-
346
- ---
347
-
348
- ## Pitfalls
349
-
350
- {From Agent 3 - Pitfalls research}
351
-
352
- ### Critical
353
- {List with prevention strategies}
354
-
355
- ### Moderate
356
- {List with prevention strategies}
357
-
358
- ### Task-Specific Warnings
359
- | When Implementing | Watch Out For | Prevention |
360
- |-------------------|---------------|------------|
361
- | ... | ... | ... |
362
-
363
- ---
364
-
365
- ## Confidence Assessment
366
-
367
- | Area | Level | Reason |
368
- |------|-------|--------|
369
- | Codebase integration | {level} | {reason} |
370
- | Implementation approach | {level} | {reason} |
371
- | Pitfalls | {level} | {reason} |
372
-
373
- ## Open Questions
374
-
375
- {Anything that couldn't be resolved - planner should be aware}
376
-
377
- ## Sources
378
-
379
- ### Codebase (from Explore)
380
- - {file references used}
381
-
382
- ### External (verified)
383
- - {Context7 queries}
384
- - {Official docs URLs}
385
-
386
- ### Community (for awareness)
387
- - {WebSearch findings - lower confidence}
388
- ```
389
-
390
- ## Phase 6: Update DECISIONS.md
391
-
392
- Any technology choices or pattern decisions from research should be recorded:
393
-
394
- ```markdown
395
- ### DEC-XXX: {Decision from research}
396
- **Date:** {today}
397
- **Status:** Active
398
- **Context:** Identified during feature research
399
- **Decision:** {what was decided}
400
- **Rationale:** {from research findings}
401
- **Implications:** {what this means for implementation}
402
- **References:** {sources}
403
- ```
404
-
405
- ## Phase 7: Commit and Complete
406
-
407
- ```bash
408
- git add .specd/features/{name}/RESEARCH.md .specd/features/{name}/DECISIONS.md
409
- git commit -m "docs({feature}): complete feature research
410
-
411
- Dimensions researched:
412
- - Codebase integration
413
- - External patterns
414
- - Pitfalls
415
-
416
- Key findings:
417
- - {one-liner from summary}"
418
- ```
419
-
420
- Present completion:
421
- ```
422
- ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
423
- RESEARCH COMPLETE ✓
424
- ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
425
-
426
- **Feature:** {name}
427
- **Confidence:** {level}
428
-
429
- ## Key Findings
430
-
431
- {3-5 bullet points}
432
-
433
- ## Files Created
434
-
435
- - `.specd/features/{name}/RESEARCH.md`
436
- - Updated `DECISIONS.md` with {N} new decisions
437
-
438
- ───────────────────────────────────────────────────────
439
-
440
- ## Next Steps
441
-
442
- /specd:feature:plan {name} — Create roadmap with phase overview
443
-
444
- <sub>/clear first — fresh context window</sub>
445
- ```
446
-
447
- </process>
448
-
449
- <success_criteria>
450
- - [ ] Feature validated (exists, has FEATURE.md)
451
- - [ ] Feature context loaded (requirements, decisions)
452
- - [ ] Codebase research completed (Explore agent)
453
- - [ ] External patterns researched (with verification)
454
- - [ ] Pitfalls catalogued (with prevention strategies)
455
- - [ ] Results synthesized into RESEARCH.md
456
- - [ ] Decisions recorded in DECISIONS.md
457
- - [ ] Files committed
458
- - [ ] User knows next step is `/specd:feature:plan`
459
- </success_criteria>
@@ -1,68 +0,0 @@
1
- ---
2
- name: specd:phase:execute
3
- description: Execute a feature plan with progress tracking
4
- argument-hint: "[feature-name] [plan-path]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Edit
9
- - Bash
10
- - Glob
11
- - Grep
12
- - AskUserQuestion
13
- ---
14
-
15
- <objective>
16
- Execute a plan from a feature, tracking progress and logging deviations.
17
-
18
- **What it does:**
19
- 1. Load context — STATE.md, DECISIONS.md, RESEARCH.md, codebase patterns
20
- 2. Find next plan — First incomplete plan, or accept path as argument
21
- 3. Execute tasks with:
22
- - Auto-fix blockers/bugs -> log to CHANGELOG.md
23
- - Ask about architectural changes -> wait for user
24
- - Run verification after each task
25
- - Stop on verification failure -> ask user (retry/skip/stop)
26
- - Commit after each task
27
- 4. Update progress — STATE.md with completed tasks
28
- 5. Suggest next — Next plan to execute
29
- </objective>
30
-
31
- <execution_context>
32
- @~/.claude/specdacular/workflows/execute-plan.md
33
- </execution_context>
34
-
35
- <context>
36
- Feature name: $ARGUMENTS (first argument)
37
- Plan path: $ARGUMENTS (optional second argument)
38
-
39
- **Load ALL feature context:**
40
- @.specd/features/{name}/STATE.md — Progress tracking
41
- @.specd/features/{name}/DECISIONS.md — Constraints to follow
42
- @.specd/features/{name}/RESEARCH.md — Implementation notes (if exists)
43
- @.specd/features/{name}/ROADMAP.md — Phase overview
44
-
45
- **Load codebase context:**
46
- @.specd/codebase/PATTERNS.md — Code patterns to follow
47
- @.specd/codebase/STRUCTURE.md — Where files go
48
- @.specd/codebase/MAP.md — System overview
49
- </context>
50
-
51
- <process>
52
- 1. **Validate** — Check feature exists with plans
53
- 2. **Load Context** — Read all feature and codebase docs
54
- 3. **Find Plan** — Next incomplete or specified plan
55
- 4. **Execute Tasks** — With verification and deviation handling
56
- 5. **Complete Plan** — Update STATE.md, suggest next
57
- </process>
58
-
59
- <success_criteria>
60
- - [ ] Feature validated with plans
61
- - [ ] Context loaded (feature, codebase)
62
- - [ ] Tasks executed in order
63
- - [ ] Verification run after each task
64
- - [ ] Deviations logged to CHANGELOG.md
65
- - [ ] STATE.md updated with progress
66
- - [ ] Commits made after each task
67
- - [ ] Next plan suggested
68
- </success_criteria>
@@ -1,62 +0,0 @@
1
- ---
2
- name: specd:phase:insert
3
- description: Insert a new phase after an existing one (decimal numbering)
4
- argument-hint: "[feature-name] [after-phase] [description...]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Edit
9
- - Bash
10
- - Glob
11
- - Grep
12
- ---
13
-
14
- <objective>
15
- Insert a new phase after an existing one using decimal numbering (e.g., Phase 03.1 after Phase 03).
16
-
17
- **What it does:**
18
- 1. Parse arguments — feature name, target phase number, description
19
- 2. Validate — Feature exists, ROADMAP.md exists, target phase exists
20
- 3. Find next decimal — Scan existing decimals, calculate next (03.1, 03.2, etc.)
21
- 4. Create phase directory — `plans/phase-{NN.M}/`
22
- 5. Update ROADMAP.md — Insert new phase section with `(INSERTED)` marker
23
- 6. Update STATE.md — Add roadmap evolution note, add unchecked phase checkbox
24
- 7. Update config.json — Increment `phases_count`
25
- 8. Show next steps
26
- </objective>
27
-
28
- <execution_context>
29
- @~/.claude/specdacular/workflows/insert-phase.md
30
- </execution_context>
31
-
32
- <context>
33
- Arguments: $ARGUMENTS (expects: feature-name after-phase description...)
34
-
35
- **Load feature context:**
36
- @.specd/features/{name}/ROADMAP.md
37
- @.specd/features/{name}/STATE.md
38
- @.specd/features/{name}/config.json
39
- </context>
40
-
41
- <process>
42
- 1. **Parse Arguments** — Extract feature-name, after-phase (integer), description
43
- 2. **Validate** — Feature exists, ROADMAP.md exists, target phase exists
44
- 3. **Find Next Decimal** — Scan for existing decimal directories, calculate next
45
- 4. **Create Phase Directory** — `plans/phase-{NN.M}/`
46
- 5. **Update ROADMAP.md** — Insert after target phase with `(INSERTED)` marker
47
- 6. **Update STATE.md** — Add evolution note and unchecked checkbox
48
- 7. **Update config.json** — Increment `phases_count`
49
- 8. **Completion** — Show summary with next steps
50
- </process>
51
-
52
- <success_criteria>
53
- - [ ] Arguments parsed correctly (feature, phase number, description)
54
- - [ ] Feature and target phase validated
55
- - [ ] Decimal number calculated correctly (based on existing decimals)
56
- - [ ] Phase directory created: `plans/phase-{NN.M}/`
57
- - [ ] ROADMAP.md updated with new phase entry (includes `(INSERTED)` marker)
58
- - [ ] Phase inserted in correct position (after target phase, before next phase)
59
- - [ ] STATE.md updated with roadmap evolution note and unchecked checkbox
60
- - [ ] config.json `phases_count` incremented
61
- - [ ] User informed of next steps
62
- </success_criteria>
@@ -1,73 +0,0 @@
1
- ---
2
- name: specd:phase:plan
3
- description: Create detailed PLAN.md files for one phase
4
- argument-hint: "[feature-name] [phase-number]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Bash
9
- - Glob
10
- - Grep
11
- - AskUserQuestion
12
- ---
13
-
14
- <objective>
15
- Create detailed, executable PLAN.md files for a single phase. Each plan is a prompt for an implementing agent.
16
-
17
- **Creates:**
18
- - `.specd/features/{name}/plans/phase-{NN}/{NN}-PLAN.md` — Executable task plans
19
-
20
- **Each PLAN.md is a prompt** for an implementing agent with:
21
- - Files to create/modify with specific paths
22
- - Code patterns to follow (from codebase)
23
- - Verification commands
24
- - Clear completion criteria
25
-
26
- **Prerequisite:** Phase should exist in ROADMAP.md. Phase preparation (`/specd:phase:prepare`) recommended but optional.
27
- </objective>
28
-
29
- <execution_context>
30
- @~/.claude/specdacular/workflows/plan-phase.md
31
- </execution_context>
32
-
33
- <context>
34
- Arguments: $ARGUMENTS (expects: feature-name phase-number)
35
-
36
- **Load feature context:**
37
- @.specd/features/{name}/FEATURE.md — Technical requirements
38
- @.specd/features/{name}/CONTEXT.md — Resolved gray areas
39
- @.specd/features/{name}/DECISIONS.md — Implementation decisions
40
- @.specd/features/{name}/RESEARCH.md — Research findings (if exists)
41
- @.specd/features/{name}/ROADMAP.md — Phase overview
42
-
43
- **Load phase context:**
44
- @.specd/features/{name}/plans/phase-{NN}/CONTEXT.md (if exists, from phase:prepare)
45
- @.specd/features/{name}/plans/phase-{NN}/RESEARCH.md (if exists, from phase:research)
46
-
47
- **Load codebase context:**
48
- @.specd/codebase/PATTERNS.md — Code patterns to follow
49
- @.specd/codebase/STRUCTURE.md — Where files go
50
- @.specd/codebase/MAP.md — System overview
51
- </context>
52
-
53
- <process>
54
- 1. **Validate** — Feature exists, phase exists, phase not already executed
55
- 2. **Load Context** — Read ALL feature, phase, and codebase docs
56
- 3. **Check Existing Plans** — If plans exist, show them and confirm replace
57
- 4. **Break Into Tasks** — 2-3 tasks per plan, sized for agent execution
58
- 5. **Write PLAN Files** — As prompts for implementing agent
59
- 6. **Update ROADMAP** — Mark this phase as planned
60
- 7. **Update STATE.md** — Record planning for this phase
61
- 8. **Commit and Present** — Points to `/specd:phase:execute`
62
- </process>
63
-
64
- <success_criteria>
65
- - [ ] Feature and phase validated
66
- - [ ] Phase not already executed
67
- - [ ] All context loaded (feature, phase, codebase)
68
- - [ ] Tasks are specific (files, actions, verification)
69
- - [ ] PLAN.md files are self-contained prompts
70
- - [ ] ROADMAP.md updated for this phase
71
- - [ ] Committed to git
72
- - [ ] User knows next step: `/specd:phase:execute`
73
- </success_criteria>