superkit-mcp-server 1.2.4 → 1.2.6

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 (162) hide show
  1. package/ARCHITECTURE.md +102 -102
  2. package/README.md +71 -71
  3. package/SUPERKIT.md +168 -168
  4. package/agents/code-archaeologist.md +106 -106
  5. package/agents/coder.md +90 -90
  6. package/agents/data-engineer.md +28 -28
  7. package/agents/devops-engineer.md +242 -242
  8. package/agents/git-manager.md +203 -203
  9. package/agents/orchestrator.md +420 -420
  10. package/agents/penetration-tester.md +188 -188
  11. package/agents/performance-optimizer.md +187 -187
  12. package/agents/planner.md +270 -270
  13. package/agents/qa-automation-engineer.md +103 -103
  14. package/agents/quant-developer.md +32 -32
  15. package/agents/reviewer.md +100 -100
  16. package/agents/scout.md +222 -222
  17. package/agents/tester.md +274 -274
  18. package/agents/ui-designer.md +208 -208
  19. package/build/__tests__/test_apply_prompt_args.js +104 -0
  20. package/build/index.js +106 -45
  21. package/build/tools/todoTools.js +39 -39
  22. package/build/tools/validators/__tests__/apiSchema.test.js +23 -23
  23. package/build/tools/validators/__tests__/convertRules.test.js +5 -5
  24. package/build/tools/validators/__tests__/frontendDesign.test.js +12 -12
  25. package/build/tools/validators/__tests__/geoChecker.test.js +19 -19
  26. package/build/tools/validators/__tests__/mobileAudit.test.js +12 -12
  27. package/build/tools/validators/__tests__/reactPerformanceChecker.test.js +17 -17
  28. package/build/tools/validators/__tests__/securityScan.test.js +6 -6
  29. package/build/tools/validators/__tests__/seoChecker.test.js +16 -16
  30. package/build/tools/validators/__tests__/typeCoverage.test.js +14 -14
  31. package/commands/README.md +122 -122
  32. package/commands/ask.toml +72 -72
  33. package/commands/brainstorm.toml +119 -119
  34. package/commands/chat.toml +77 -77
  35. package/commands/code-preview.toml +37 -37
  36. package/commands/code.toml +28 -28
  37. package/commands/content.toml +200 -200
  38. package/commands/cook.toml +77 -77
  39. package/commands/copywrite.toml +131 -131
  40. package/commands/db.toml +192 -192
  41. package/commands/debug.toml +166 -166
  42. package/commands/design.toml +158 -158
  43. package/commands/dev-rules.toml +14 -14
  44. package/commands/do.toml +117 -117
  45. package/commands/doc-rules.toml +14 -14
  46. package/commands/docs.toml +148 -148
  47. package/commands/fix.toml +440 -440
  48. package/commands/fullstack.toml +175 -175
  49. package/commands/git.toml +235 -235
  50. package/commands/help.toml +84 -84
  51. package/commands/integrate.toml +127 -127
  52. package/commands/journal.toml +136 -136
  53. package/commands/kit-setup.toml +40 -40
  54. package/commands/mcp.toml +183 -183
  55. package/commands/orchestration.toml +15 -15
  56. package/commands/plan.toml +206 -172
  57. package/commands/pm.toml +148 -148
  58. package/commands/pr.toml +50 -50
  59. package/commands/project.toml +32 -32
  60. package/commands/research.toml +117 -117
  61. package/commands/review-pr.toml +63 -63
  62. package/commands/review.toml +190 -190
  63. package/commands/scout-ext.toml +97 -97
  64. package/commands/scout.toml +79 -79
  65. package/commands/screenshot.toml +65 -65
  66. package/commands/session.toml +102 -102
  67. package/commands/skill.toml +384 -384
  68. package/commands/status.toml +22 -22
  69. package/commands/team.toml +56 -56
  70. package/commands/test.toml +164 -164
  71. package/commands/ticket.toml +70 -70
  72. package/commands/use.toml +106 -106
  73. package/commands/video.toml +83 -83
  74. package/commands/watzup.toml +71 -71
  75. package/commands/workflow.toml +14 -14
  76. package/package.json +35 -35
  77. package/skills/meta/README.md +30 -30
  78. package/skills/meta/api-design/SKILL.md +134 -134
  79. package/skills/meta/code-review/SKILL.md +44 -44
  80. package/skills/meta/code-review/checklists/pre-merge.md +25 -25
  81. package/skills/meta/code-review/workflows/architecture-pass.md +26 -26
  82. package/skills/meta/code-review/workflows/performance-pass.md +27 -27
  83. package/skills/meta/code-review/workflows/security-pass.md +29 -29
  84. package/skills/meta/compound-docs/SKILL.md +133 -133
  85. package/skills/meta/debug/SKILL.md +40 -40
  86. package/skills/meta/debug/templates/bug-report.template.md +31 -31
  87. package/skills/meta/debug/workflows/reproduce-issue.md +20 -20
  88. package/skills/meta/docker/SKILL.md +126 -126
  89. package/skills/meta/examples/supabase/SKILL.md +46 -46
  90. package/skills/meta/examples/supabase/references/best-practices.md +319 -319
  91. package/skills/meta/examples/supabase/references/common-patterns.md +373 -373
  92. package/skills/meta/examples/supabase/templates/migration-template.sql +49 -49
  93. package/skills/meta/examples/supabase/templates/rls-policy-template.sql +77 -77
  94. package/skills/meta/examples/supabase/workflows/debugging.md +260 -260
  95. package/skills/meta/examples/supabase/workflows/migration-workflow.md +211 -211
  96. package/skills/meta/examples/supabase/workflows/rls-policies.md +244 -244
  97. package/skills/meta/examples/supabase/workflows/schema-design.md +321 -321
  98. package/skills/meta/file-todos/SKILL.md +88 -88
  99. package/skills/meta/mobile/SKILL.md +140 -140
  100. package/skills/meta/nextjs/SKILL.md +101 -101
  101. package/skills/meta/performance/SKILL.md +130 -130
  102. package/skills/meta/react-patterns/SKILL.md +83 -83
  103. package/skills/meta/security/SKILL.md +114 -114
  104. package/skills/meta/session-resume/SKILL.md +96 -96
  105. package/skills/meta/tailwind/SKILL.md +139 -139
  106. package/skills/meta/testing/SKILL.md +43 -43
  107. package/skills/meta/testing/references/vitest-patterns.md +45 -45
  108. package/skills/meta/testing/templates/component-test.template.tsx +37 -37
  109. package/skills/tech/alpha-vantage/SKILL.md +142 -142
  110. package/skills/tech/alpha-vantage/references/commodities.md +153 -153
  111. package/skills/tech/alpha-vantage/references/economic-indicators.md +158 -158
  112. package/skills/tech/alpha-vantage/references/forex-crypto.md +154 -154
  113. package/skills/tech/alpha-vantage/references/fundamentals.md +223 -223
  114. package/skills/tech/alpha-vantage/references/intelligence.md +138 -138
  115. package/skills/tech/alpha-vantage/references/options.md +93 -93
  116. package/skills/tech/alpha-vantage/references/technical-indicators.md +374 -374
  117. package/skills/tech/alpha-vantage/references/time-series.md +157 -157
  118. package/skills/tech/financial-modeling/SKILL.md +18 -18
  119. package/skills/tech/financial-modeling/skills/3-statements/SKILL.md +368 -368
  120. package/skills/tech/financial-modeling/skills/3-statements/references/formatting.md +118 -118
  121. package/skills/tech/financial-modeling/skills/3-statements/references/formulas.md +292 -292
  122. package/skills/tech/financial-modeling/skills/3-statements/references/sec-filings.md +125 -125
  123. package/skills/tech/financial-modeling/skills/dcf-model/SKILL.md +1210 -1210
  124. package/skills/tech/financial-modeling/skills/dcf-model/TROUBLESHOOTING.md +40 -40
  125. package/skills/tech/financial-modeling/skills/dcf-model/requirements.txt +8 -8
  126. package/skills/tech/financial-modeling/skills/dcf-model/scripts/validate_dcf.py +292 -292
  127. package/skills/tech/financial-modeling/skills/lbo-model/SKILL.md +236 -236
  128. package/skills/tech/financial-modeling/skills/merger-model/SKILL.md +108 -108
  129. package/skills/workflows/README.md +203 -203
  130. package/skills/workflows/adr.md +174 -174
  131. package/skills/workflows/changelog.md +74 -74
  132. package/skills/workflows/compound.md +323 -323
  133. package/skills/workflows/compound_health.md +74 -74
  134. package/skills/workflows/create-agent-skill.md +138 -138
  135. package/skills/workflows/cycle.md +144 -144
  136. package/skills/workflows/deploy-docs.md +84 -84
  137. package/skills/workflows/development-rules.md +42 -42
  138. package/skills/workflows/doc.md +95 -95
  139. package/skills/workflows/documentation-management.md +34 -34
  140. package/skills/workflows/explore.md +146 -146
  141. package/skills/workflows/generate_command.md +106 -106
  142. package/skills/workflows/heal-skill.md +97 -97
  143. package/skills/workflows/housekeeping.md +229 -229
  144. package/skills/workflows/kit-setup.md +102 -102
  145. package/skills/workflows/map-codebase.md +78 -78
  146. package/skills/workflows/orchestration-protocol.md +43 -43
  147. package/skills/workflows/plan-compound.md +439 -439
  148. package/skills/workflows/plan_review.md +269 -269
  149. package/skills/workflows/primary-workflow.md +37 -37
  150. package/skills/workflows/promote_pattern.md +86 -86
  151. package/skills/workflows/release-docs.md +82 -82
  152. package/skills/workflows/report-bug.md +135 -135
  153. package/skills/workflows/reproduce-bug.md +118 -118
  154. package/skills/workflows/resolve_pr.md +133 -133
  155. package/skills/workflows/resolve_todo.md +128 -128
  156. package/skills/workflows/review-compound.md +376 -376
  157. package/skills/workflows/skill-review.md +127 -127
  158. package/skills/workflows/specs.md +257 -257
  159. package/skills/workflows/triage-sprint.md +102 -102
  160. package/skills/workflows/triage.md +152 -152
  161. package/skills/workflows/work.md +399 -399
  162. package/skills/workflows/xcode-test.md +93 -93
@@ -1,323 +1,323 @@
1
- ---
2
- description: Document reusable knowledge (solutions, features, decisions) to compound project capability.
3
- ---
4
-
5
- # /compound - Compound Your Knowledge
6
-
7
- Capture reusable knowledge while context is fresh—solutions to problems, feature implementations, architectural decisions—creating structured documentation for searchability and future reference.
8
-
9
- > **Why "compound"?** Each documented piece of knowledge compounds your capability. The first time you solve a problem OR build a feature takes research. Document it, and the next occurrence takes minutes.
10
-
11
- ## When To Use
12
-
13
- **Auto-trigger on phrases:**
14
- - "that worked"
15
- - "it's fixed"
16
- - "working now"
17
- - "problem solved"
18
-
19
- **Also trigger when completing features:**
20
- - "Feature complete"
21
- - "Shipped the {feature}"
22
- - "Implementation done"
23
-
24
- **Manual invocation:**
25
- - `/compound` after solving a non-trivial problem
26
- - `/compound` after completing a feature
27
-
28
- **Skip documentation for:**
29
- - Simple typos
30
- - Obvious syntax errors
31
- - Trivial fixes (<5 minutes to solve)
32
-
33
- ---
34
-
35
- ## Workflow
36
-
37
- ### Step 0: Instrumentation
38
-
39
- ```bash
40
- Call MCP `call_tool_logger_manager` { action: "logWorkflow", name: "/compound", outcome: "success" }
41
- ```
42
-
43
- ### Step 0.5: Search for Similar Solutions (Recommended)
44
-
45
- > [!TIP]
46
- > Check if a similar solution already exists to avoid duplication. Search for keywords related to the problem or feature you are documenting.
47
-
48
- ```bash
49
- Call MCP `call_tool_compound_manager` { action: "search", terms: [ "{primary symptom or feature keywords}"] }
50
- ```
51
- ```
52
-
53
- ### Step 1: Gather Context from Conversation
54
-
55
- Extract the following from the current conversation:
56
-
57
- **Required:**
58
- - [ ] **Problem/Feature:** What was solved or built?
59
- - [ ] **Details:** Root cause (for problems) or Implementation details (for features)
60
- - [ ] **Solution:** What fixed it? (if problem)
61
-
62
- **Optional but valuable:**
63
- - [ ] **Investigation attempts:** What didn't work and why?
64
- - [ ] **Prevention strategies:** How to avoid in future?
65
- - [ ] **Related issues:** Similar problems encountered before?
66
-
67
- ### Step 2: Determine Documentation Type
68
-
69
- | If you are documenting... | Output |
70
- |---------------------------|--------|
71
- | A solved problem | Create in `docs/solutions/{category}/` |
72
- | A new feature | Create in `docs/features/` OR update README |
73
- | An architectural decision | Create in `docs/decisions/` (ADR) |
74
-
75
- ### Step 3: Determine Details (if Problem)
76
-
77
- If documenting a **Solved Problem**, map to one of these categories:
78
-
79
- | If the problem involves... | Category |
80
- |---------------------------|----------|
81
- | Slow performance, N+1 queries, memory | `performance-issues` |
82
- | Vulnerabilities, auth, data exposure | `security-issues` |
83
- | Schema, migrations, data integrity | `database-issues` |
84
- | Compilation, bundling, dependencies | `build-errors` |
85
- | Failing tests, flaky tests | `test-failures` |
86
- | Exceptions, crashes, runtime errors | `runtime-errors` |
87
- | Visual glitches, responsive, accessibility | `ui-bugs` |
88
- | API failures, third-party services | `integration-issues` |
89
- | Business logic, calculations | `logic-errors` |
90
-
91
- ### Step 4: Generate Filename
92
-
93
- Format: `{sanitized-symptom}-{YYYYMMDD}.md`
94
-
95
- **Sanitization rules:**
96
- - Lowercase
97
- - Replace spaces with hyphens
98
- - Remove special characters except hyphens
99
- - Truncate to <60 characters
100
-
101
- **Examples:**
102
- - `n-plus-one-query-user-list-20251219.md`
103
- - `auth-cookie-mismatch-20251219.md`
104
- - `missing-env-variable-supabase-20251219.md`
105
-
106
- ### Step 5: Create Solution Document (if Problem)
107
-
108
- Load the template from `docs/solutions/templates/solution-template.md` and fill in:
109
-
110
- **YAML Frontmatter:**
111
- ```yaml
112
- ---
113
- date: "{TODAY'S DATE: YYYY-MM-DD}"
114
- problem_type: "{FROM SCHEMA: performance_issue, security_issue, etc.}"
115
- component: "{FROM SCHEMA: api_endpoint, database_model, etc.}"
116
- severity: "{critical | high | medium | low}"
117
- symptoms:
118
- - "{EXACT ERROR MESSAGE OR BEHAVIOR}"
119
- root_cause: "{FROM SCHEMA: missing_include, null_check_missing, etc.}"
120
- tags:
121
- - "{relevant-tag-1}"
122
- - "{relevant-tag-2}"
123
- related_issues: []
124
- ---
125
- ```
126
-
127
- **Document Sections:**
128
- 1. Problem Statement with impact
129
- 2. Symptoms observed
130
- 3. Investigation steps with results table
131
- 4. Root cause analysis
132
- 5. Working solution with code examples
133
- 6. Prevention strategies
134
- 7. Cross-references
135
-
136
- **For Feature Documentation:**
137
- - Use `docs/features/templates/feature-template.md`
138
- - Create in `docs/features/{feature-name}.md`
139
-
140
- ### Step 6: Validate YAML
141
-
142
- Before creating the file, validate:
143
-
144
- - [ ] `problem_type` is a valid enum from schema.yaml
145
- - [ ] `severity` is one of: critical, high, medium, low
146
- - [ ] `component` is a valid enum from schema.yaml
147
- - [ ] `root_cause` is a valid enum from schema.yaml
148
- - [ ] `symptoms` is a non-empty array
149
- - [ ] `tags` is a non-empty array
150
-
151
- If validation fails, fix the values before proceeding.
152
-
153
- ### Step 7: Write Solution File
154
-
155
- Create the file at: `docs/solutions/{category}/{filename}.md`
156
-
157
- ### Step 8: Check for Pattern Promotion
158
-
159
- After creating the solution, **dynamically count** similar solutions:
160
-
161
- ```bash
162
- # Count similar solutions
163
- grep -l "{key symptom terms}" docs/solutions/**/*.md | wc -l
164
- ```
165
-
166
- **If count ≥ 3:**
167
- 1. Create a pattern promotion todo with a **DRAFT**:
168
-
169
- ```bash
170
- # Create todo for pattern promotion
171
- Call MCP `call_tool_todo_manager` { action: "create", priority: "p2", title: "Promote Pattern: {Pattern Name}", description: "TODO description" }
172
- "Pattern identified during solution documentation: {Pattern Name}.\n\nContext:\n(Description of the pattern and why it matters)" \
173
- "Draft the pattern in docs/solutions/patterns/critical-patterns.md" \
174
- "Add 'WRONG' vs 'CORRECT' examples" \
175
- "Link to original solution: ${filename}"
176
- ```
177
-
178
- 2. The todo should include:
179
- - Pattern name
180
- - List of 3+ related solutions
181
- - Proposed ❌ WRONG vs ✅ CORRECT format
182
- - Source cross-references
183
-
184
- ### Step 9: Capture Follow-up Opportunities (Conditional)
185
-
186
- > [!TIP]
187
- > **Safety Net.** If you already ran `/review` and captured deferred work there, **SKIP THIS STEP.**
188
-
189
- **If running standalone (no prior review), check:**
190
- - [ ] Did the solution defer any enhancements for later?
191
- - [ ] Are there related optimizations not yet implemented?
192
- - [ ] Did we close/reject alternatives that have value to port later?
193
-
194
- **Also check for NEW opportunities (discovered during documentation):**
195
- - [ ] Did writing this solution reveal a gap in another area?
196
-
197
- **If YES, create todo files:**
198
-
199
- ```bash
200
- # Get next ID
201
- Call MCP `call_tool_todo_manager` { action: "nextId" }
202
-
203
- # Create todo from template
204
- cp todos/todo-template.md todos/${next_id}-pending-p2-{description}.md
205
- # Edit the file with specific details of deferred tasks
206
- ```
207
-
208
- > [!TIP]
209
- > Use the standardized naming convention: `{id}-{status}-{priority}-{description}.md`.
210
- > See [todos/README.md](../../todos/README.md) for details.
211
- ```
212
-
213
- ### Step 10: Mandatory Validation
214
-
215
- > [!CAUTION]
216
- > **Do not skip.** Auto-validation protects the knowledge base.
217
-
218
- Before closing the `/compound` workflow, run:
219
-
220
- ```bash
221
- Call MCP `call_tool_compound_manager` { action: "validate" }
222
- ```
223
-
224
- - [ ] Script passed?
225
- - [ ] Pattern promotion todo created (if count ≥ 3)?
226
- - [ ] Deferred work converted to todos?
227
-
228
- **If any failure, FIX IT before closing.**
229
-
230
- ---
231
-
232
- ### Phase 5: Completion & Handoff
233
-
234
- #### Step 1: Establish Terminal UI State
235
-
236
- > [!IMPORTANT]
237
- > **Visual Completion Signal**
238
- > Call `task_boundary` one last time to signal completion in the user's UI. This prevents the "task" from appearing active after you've finished.
239
-
240
- ```javascript
241
- await task_boundary({
242
- TaskName: "[COMPLETED] Compound Knowledge: {filename}",
243
- TaskStatus: "Solution documented and validated. Offering next steps.",
244
- Mode: "VERIFICATION",
245
- TaskSummary: "Documented {type} at docs/solutions/{category}/{filename}.md. Validated YAML and pattern promotion."
246
- });
247
- ```
248
-
249
- #### Step 2: Mandatory Handoff
250
-
251
- > [!IMPORTANT]
252
- > **Exit Transition**
253
- > Do not stop here. Offer the user clear paths to the next logical workflow.
254
-
255
- ```bash
256
- ✓ Solution documented
257
-
258
- File created:
259
- - docs/solutions/{category}/{filename}.md
260
-
261
- Next steps:
262
- 1. /housekeeping - Archive and cleanup before push (Recommended)
263
- 2. /review - Get feedback on documentation or associated changes
264
- 3. /work - Return to previous task or start new implementation
265
- 4. Continue working - Perform manual follow-ups
266
- ```
267
-
268
- ---
269
-
270
- ## Quality Guidelines
271
-
272
- **Good documentation has:**
273
- - ✅ Exact error messages (copy-paste from output)
274
- - ✅ Specific file:line references
275
- - ✅ Observable symptoms (what you saw, not interpretations)
276
- - ✅ Failed attempts documented (helps avoid wrong paths)
277
- - ✅ Technical explanation (not just "what" but "why")
278
- - ✅ Code examples (before/after if applicable)
279
- - ✅ Prevention guidance (how to catch early)
280
-
281
- **Avoid:**
282
- - ❌ Vague descriptions ("something was wrong")
283
- - ❌ Missing technical details ("fixed the code")
284
- - ❌ No context (which file? which version?)
285
- - ❌ Just code dumps (explain why it works)
286
-
287
- ---
288
-
289
- ## Example
290
-
291
- **User:** "That worked! The N+1 query is fixed."
292
-
293
- **Agent creates:**
294
-
295
- ```
296
- docs/solutions/performance-issues/n-plus-one-user-list-20251219.md
297
- ```
298
-
299
- With frontmatter:
300
- ```yaml
301
- ---
302
- date: "2025-12-19"
303
- problem_type: "performance_issue"
304
- component: "database_model"
305
- severity: "high"
306
- symptoms:
307
- - "User list API taking >5 seconds"
308
- - "N+1 query detected in logs"
309
- root_cause: "missing_include"
310
- tags:
311
- - n-plus-one
312
- - eager-loading
313
- - performance
314
- ---
315
- ```
316
-
317
- ---
318
-
319
- ## References
320
-
321
- - Schema: `docs/solutions/schema.yaml`
322
- - Template: `docs/solutions/templates/solution-template.md`
323
- - Patterns: `docs/solutions/patterns/critical-patterns.md`
1
+ ---
2
+ description: Document reusable knowledge (solutions, features, decisions) to compound project capability.
3
+ ---
4
+
5
+ # /compound - Compound Your Knowledge
6
+
7
+ Capture reusable knowledge while context is fresh—solutions to problems, feature implementations, architectural decisions—creating structured documentation for searchability and future reference.
8
+
9
+ > **Why "compound"?** Each documented piece of knowledge compounds your capability. The first time you solve a problem OR build a feature takes research. Document it, and the next occurrence takes minutes.
10
+
11
+ ## When To Use
12
+
13
+ **Auto-trigger on phrases:**
14
+ - "that worked"
15
+ - "it's fixed"
16
+ - "working now"
17
+ - "problem solved"
18
+
19
+ **Also trigger when completing features:**
20
+ - "Feature complete"
21
+ - "Shipped the {feature}"
22
+ - "Implementation done"
23
+
24
+ **Manual invocation:**
25
+ - `/compound` after solving a non-trivial problem
26
+ - `/compound` after completing a feature
27
+
28
+ **Skip documentation for:**
29
+ - Simple typos
30
+ - Obvious syntax errors
31
+ - Trivial fixes (<5 minutes to solve)
32
+
33
+ ---
34
+
35
+ ## Workflow
36
+
37
+ ### Step 0: Instrumentation
38
+
39
+ ```bash
40
+ Call MCP `call_tool_logger_manager` { action: "logWorkflow", name: "/compound", outcome: "success" }
41
+ ```
42
+
43
+ ### Step 0.5: Search for Similar Solutions (Recommended)
44
+
45
+ > [!TIP]
46
+ > Check if a similar solution already exists to avoid duplication. Search for keywords related to the problem or feature you are documenting.
47
+
48
+ ```bash
49
+ Call MCP `call_tool_compound_manager` { action: "search", terms: [ "{primary symptom or feature keywords}"] }
50
+ ```
51
+ ```
52
+
53
+ ### Step 1: Gather Context from Conversation
54
+
55
+ Extract the following from the current conversation:
56
+
57
+ **Required:**
58
+ - [ ] **Problem/Feature:** What was solved or built?
59
+ - [ ] **Details:** Root cause (for problems) or Implementation details (for features)
60
+ - [ ] **Solution:** What fixed it? (if problem)
61
+
62
+ **Optional but valuable:**
63
+ - [ ] **Investigation attempts:** What didn't work and why?
64
+ - [ ] **Prevention strategies:** How to avoid in future?
65
+ - [ ] **Related issues:** Similar problems encountered before?
66
+
67
+ ### Step 2: Determine Documentation Type
68
+
69
+ | If you are documenting... | Output |
70
+ |---------------------------|--------|
71
+ | A solved problem | Create in `docs/solutions/{category}/` |
72
+ | A new feature | Create in `docs/features/` OR update README |
73
+ | An architectural decision | Create in `docs/decisions/` (ADR) |
74
+
75
+ ### Step 3: Determine Details (if Problem)
76
+
77
+ If documenting a **Solved Problem**, map to one of these categories:
78
+
79
+ | If the problem involves... | Category |
80
+ |---------------------------|----------|
81
+ | Slow performance, N+1 queries, memory | `performance-issues` |
82
+ | Vulnerabilities, auth, data exposure | `security-issues` |
83
+ | Schema, migrations, data integrity | `database-issues` |
84
+ | Compilation, bundling, dependencies | `build-errors` |
85
+ | Failing tests, flaky tests | `test-failures` |
86
+ | Exceptions, crashes, runtime errors | `runtime-errors` |
87
+ | Visual glitches, responsive, accessibility | `ui-bugs` |
88
+ | API failures, third-party services | `integration-issues` |
89
+ | Business logic, calculations | `logic-errors` |
90
+
91
+ ### Step 4: Generate Filename
92
+
93
+ Format: `{sanitized-symptom}-{YYYYMMDD}.md`
94
+
95
+ **Sanitization rules:**
96
+ - Lowercase
97
+ - Replace spaces with hyphens
98
+ - Remove special characters except hyphens
99
+ - Truncate to <60 characters
100
+
101
+ **Examples:**
102
+ - `n-plus-one-query-user-list-20251219.md`
103
+ - `auth-cookie-mismatch-20251219.md`
104
+ - `missing-env-variable-supabase-20251219.md`
105
+
106
+ ### Step 5: Create Solution Document (if Problem)
107
+
108
+ Load the template from `docs/solutions/templates/solution-template.md` and fill in:
109
+
110
+ **YAML Frontmatter:**
111
+ ```yaml
112
+ ---
113
+ date: "{TODAY'S DATE: YYYY-MM-DD}"
114
+ problem_type: "{FROM SCHEMA: performance_issue, security_issue, etc.}"
115
+ component: "{FROM SCHEMA: api_endpoint, database_model, etc.}"
116
+ severity: "{critical | high | medium | low}"
117
+ symptoms:
118
+ - "{EXACT ERROR MESSAGE OR BEHAVIOR}"
119
+ root_cause: "{FROM SCHEMA: missing_include, null_check_missing, etc.}"
120
+ tags:
121
+ - "{relevant-tag-1}"
122
+ - "{relevant-tag-2}"
123
+ related_issues: []
124
+ ---
125
+ ```
126
+
127
+ **Document Sections:**
128
+ 1. Problem Statement with impact
129
+ 2. Symptoms observed
130
+ 3. Investigation steps with results table
131
+ 4. Root cause analysis
132
+ 5. Working solution with code examples
133
+ 6. Prevention strategies
134
+ 7. Cross-references
135
+
136
+ **For Feature Documentation:**
137
+ - Use `docs/features/templates/feature-template.md`
138
+ - Create in `docs/features/{feature-name}.md`
139
+
140
+ ### Step 6: Validate YAML
141
+
142
+ Before creating the file, validate:
143
+
144
+ - [ ] `problem_type` is a valid enum from schema.yaml
145
+ - [ ] `severity` is one of: critical, high, medium, low
146
+ - [ ] `component` is a valid enum from schema.yaml
147
+ - [ ] `root_cause` is a valid enum from schema.yaml
148
+ - [ ] `symptoms` is a non-empty array
149
+ - [ ] `tags` is a non-empty array
150
+
151
+ If validation fails, fix the values before proceeding.
152
+
153
+ ### Step 7: Write Solution File
154
+
155
+ Create the file at: `docs/solutions/{category}/{filename}.md`
156
+
157
+ ### Step 8: Check for Pattern Promotion
158
+
159
+ After creating the solution, **dynamically count** similar solutions:
160
+
161
+ ```bash
162
+ # Count similar solutions
163
+ grep -l "{key symptom terms}" docs/solutions/**/*.md | wc -l
164
+ ```
165
+
166
+ **If count ≥ 3:**
167
+ 1. Create a pattern promotion todo with a **DRAFT**:
168
+
169
+ ```bash
170
+ # Create todo for pattern promotion
171
+ Call MCP `call_tool_todo_manager` { action: "create", priority: "p2", title: "Promote Pattern: {Pattern Name}", description: "TODO description" }
172
+ "Pattern identified during solution documentation: {Pattern Name}.\n\nContext:\n(Description of the pattern and why it matters)" \
173
+ "Draft the pattern in docs/solutions/patterns/critical-patterns.md" \
174
+ "Add 'WRONG' vs 'CORRECT' examples" \
175
+ "Link to original solution: ${filename}"
176
+ ```
177
+
178
+ 2. The todo should include:
179
+ - Pattern name
180
+ - List of 3+ related solutions
181
+ - Proposed ❌ WRONG vs ✅ CORRECT format
182
+ - Source cross-references
183
+
184
+ ### Step 9: Capture Follow-up Opportunities (Conditional)
185
+
186
+ > [!TIP]
187
+ > **Safety Net.** If you already ran `/review` and captured deferred work there, **SKIP THIS STEP.**
188
+
189
+ **If running standalone (no prior review), check:**
190
+ - [ ] Did the solution defer any enhancements for later?
191
+ - [ ] Are there related optimizations not yet implemented?
192
+ - [ ] Did we close/reject alternatives that have value to port later?
193
+
194
+ **Also check for NEW opportunities (discovered during documentation):**
195
+ - [ ] Did writing this solution reveal a gap in another area?
196
+
197
+ **If YES, create todo files:**
198
+
199
+ ```bash
200
+ # Get next ID
201
+ Call MCP `call_tool_todo_manager` { action: "nextId" }
202
+
203
+ # Create todo from template
204
+ cp todos/todo-template.md todos/${next_id}-pending-p2-{description}.md
205
+ # Edit the file with specific details of deferred tasks
206
+ ```
207
+
208
+ > [!TIP]
209
+ > Use the standardized naming convention: `{id}-{status}-{priority}-{description}.md`.
210
+ > See [todos/README.md](../../todos/README.md) for details.
211
+ ```
212
+
213
+ ### Step 10: Mandatory Validation
214
+
215
+ > [!CAUTION]
216
+ > **Do not skip.** Auto-validation protects the knowledge base.
217
+
218
+ Before closing the `/compound` workflow, run:
219
+
220
+ ```bash
221
+ Call MCP `call_tool_compound_manager` { action: "validate" }
222
+ ```
223
+
224
+ - [ ] Script passed?
225
+ - [ ] Pattern promotion todo created (if count ≥ 3)?
226
+ - [ ] Deferred work converted to todos?
227
+
228
+ **If any failure, FIX IT before closing.**
229
+
230
+ ---
231
+
232
+ ### Phase 5: Completion & Handoff
233
+
234
+ #### Step 1: Establish Terminal UI State
235
+
236
+ > [!IMPORTANT]
237
+ > **Visual Completion Signal**
238
+ > Call `task_boundary` one last time to signal completion in the user's UI. This prevents the "task" from appearing active after you've finished.
239
+
240
+ ```javascript
241
+ await task_boundary({
242
+ TaskName: "[COMPLETED] Compound Knowledge: {filename}",
243
+ TaskStatus: "Solution documented and validated. Offering next steps.",
244
+ Mode: "VERIFICATION",
245
+ TaskSummary: "Documented {type} at docs/solutions/{category}/{filename}.md. Validated YAML and pattern promotion."
246
+ });
247
+ ```
248
+
249
+ #### Step 2: Mandatory Handoff
250
+
251
+ > [!IMPORTANT]
252
+ > **Exit Transition**
253
+ > Do not stop here. Offer the user clear paths to the next logical workflow.
254
+
255
+ ```bash
256
+ ✓ Solution documented
257
+
258
+ File created:
259
+ - docs/solutions/{category}/{filename}.md
260
+
261
+ Next steps:
262
+ 1. /housekeeping - Archive and cleanup before push (Recommended)
263
+ 2. /review - Get feedback on documentation or associated changes
264
+ 3. /work - Return to previous task or start new implementation
265
+ 4. Continue working - Perform manual follow-ups
266
+ ```
267
+
268
+ ---
269
+
270
+ ## Quality Guidelines
271
+
272
+ **Good documentation has:**
273
+ - ✅ Exact error messages (copy-paste from output)
274
+ - ✅ Specific file:line references
275
+ - ✅ Observable symptoms (what you saw, not interpretations)
276
+ - ✅ Failed attempts documented (helps avoid wrong paths)
277
+ - ✅ Technical explanation (not just "what" but "why")
278
+ - ✅ Code examples (before/after if applicable)
279
+ - ✅ Prevention guidance (how to catch early)
280
+
281
+ **Avoid:**
282
+ - ❌ Vague descriptions ("something was wrong")
283
+ - ❌ Missing technical details ("fixed the code")
284
+ - ❌ No context (which file? which version?)
285
+ - ❌ Just code dumps (explain why it works)
286
+
287
+ ---
288
+
289
+ ## Example
290
+
291
+ **User:** "That worked! The N+1 query is fixed."
292
+
293
+ **Agent creates:**
294
+
295
+ ```
296
+ docs/solutions/performance-issues/n-plus-one-user-list-20251219.md
297
+ ```
298
+
299
+ With frontmatter:
300
+ ```yaml
301
+ ---
302
+ date: "2025-12-19"
303
+ problem_type: "performance_issue"
304
+ component: "database_model"
305
+ severity: "high"
306
+ symptoms:
307
+ - "User list API taking >5 seconds"
308
+ - "N+1 query detected in logs"
309
+ root_cause: "missing_include"
310
+ tags:
311
+ - n-plus-one
312
+ - eager-loading
313
+ - performance
314
+ ---
315
+ ```
316
+
317
+ ---
318
+
319
+ ## References
320
+
321
+ - Schema: `docs/solutions/schema.yaml`
322
+ - Template: `docs/solutions/templates/solution-template.md`
323
+ - Patterns: `docs/solutions/patterns/critical-patterns.md`