ai-developer-skill-os 7.0.2 → 7.5.0

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 (188) hide show
  1. package/.agents/AGENTS.md +44 -88
  2. package/.agents/CHANGELOG.md +69 -0
  3. package/.agents/LICENSE +21 -0
  4. package/.agents/README.md +59 -0
  5. package/.agents/_template/BEHAVIOR_SPEC.md +96 -0
  6. package/.agents/_template/examples/example-en.md +49 -0
  7. package/.agents/_template/examples/example-vi.md +49 -0
  8. package/.agents/docs/CHI_TIET_SKILLS.md +125 -0
  9. package/.agents/docs/GOVERNANCE.md +40 -0
  10. package/.agents/docs/HUONG_DAN_SU_DUNG.md +120 -0
  11. package/.agents/docs/SPEC.md +87 -0
  12. package/.agents/docs/adr/0001-intent-based-architecture.md +19 -0
  13. package/.agents/docs/adr/0002-kernel-freeze.md +21 -0
  14. package/.agents/docs/adr/0003-risk-based-verification.md +20 -0
  15. package/.agents/docs/adr/0004-progressive-evidence.md +19 -0
  16. package/.agents/docs/skill-classification.md +25 -0
  17. package/.agents/skills/qk-access-policy/SKILL.md +179 -0
  18. package/.agents/skills/qk-ai-builder/SKILL.md +215 -0
  19. package/.agents/skills/qk-api-lifecycle/SKILL.md +176 -0
  20. package/.agents/skills/qk-bug-resolution/SKILL.md +307 -0
  21. package/.agents/skills/qk-context-loader/SKILL.md +218 -0
  22. package/.agents/skills/qk-data-lifecycle/SKILL.md +192 -0
  23. package/.agents/skills/qk-db-optimizer/SKILL.md +196 -0
  24. package/.agents/skills/qk-design-to-code/SKILL.md +285 -0
  25. package/.agents/skills/qk-docs/SKILL.md +198 -0
  26. package/.agents/skills/qk-engineering-standard/SKILL.md +351 -0
  27. package/.agents/skills/qk-engineering-standard/rules/backend.md +122 -0
  28. package/.agents/skills/qk-engineering-standard/rules/database.md +3 -0
  29. package/.agents/skills/qk-engineering-standard/rules/frontend.md +152 -0
  30. package/.agents/skills/qk-engineering-standard/rules/security.md +3 -0
  31. package/.agents/skills/qk-engineering-standard/rules/testing.md +3 -0
  32. package/.agents/skills/qk-fe-api-integration/SKILL.md +326 -0
  33. package/.agents/skills/qk-feature-delivery/SKILL.md +305 -0
  34. package/.agents/skills/qk-help/SKILL.md +193 -0
  35. package/.agents/skills/qk-orchestrator/SKILL.md +277 -0
  36. package/.agents/skills/qk-orchestrator/references/routing-table.md +77 -0
  37. package/.agents/skills/qk-production-release/SKILL.md +284 -0
  38. package/.agents/skills/qk-project-bootstrap/SKILL.md +234 -0
  39. package/.agents/skills/qk-project-health/SKILL.md +199 -0
  40. package/.agents/skills/qk-project-memory/SKILL.md +172 -0
  41. package/.agents/skills/qk-system-evolution/SKILL.md +281 -0
  42. package/.agents/skills/qk-ui-audit/SKILL.md +314 -0
  43. package/.agents/skills/qk-ui-audit/references/anti-slop-checklist.md +136 -0
  44. package/.agents/skills/qk-ui-system-builder/SKILL.md +221 -0
  45. package/.agents/skills/qk-validation-gate/SKILL.md +359 -0
  46. package/.agents/skills.json +819 -0
  47. package/.github/workflows/ci.yml +1 -1
  48. package/.qk-ai-skill-os/CHANGELOG.md +69 -0
  49. package/.qk-ai-skill-os/LICENSE +21 -0
  50. package/.qk-ai-skill-os/README.md +59 -0
  51. package/.qk-ai-skill-os/_template/BEHAVIOR_SPEC.md +96 -0
  52. package/.qk-ai-skill-os/_template/examples/example-en.md +49 -0
  53. package/.qk-ai-skill-os/_template/examples/example-vi.md +49 -0
  54. package/.qk-ai-skill-os/docs/CHI_TIET_SKILLS.md +125 -0
  55. package/.qk-ai-skill-os/docs/GOVERNANCE.md +40 -0
  56. package/.qk-ai-skill-os/docs/HUONG_DAN_SU_DUNG.md +120 -0
  57. package/.qk-ai-skill-os/docs/SPEC.md +87 -0
  58. package/.qk-ai-skill-os/docs/adr/0001-intent-based-architecture.md +19 -0
  59. package/.qk-ai-skill-os/docs/adr/0002-kernel-freeze.md +21 -0
  60. package/.qk-ai-skill-os/docs/adr/0003-risk-based-verification.md +20 -0
  61. package/.qk-ai-skill-os/docs/adr/0004-progressive-evidence.md +19 -0
  62. package/.qk-ai-skill-os/docs/skill-classification.md +25 -0
  63. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-accessibility-audit/SKILL.md +121 -0
  64. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-agent-orchestrator/SKILL.md +179 -0
  65. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-api-integration/SKILL.md +389 -0
  66. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-auth-security/SKILL.md +96 -0
  67. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-backend-architecture/SKILL.md +125 -0
  68. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-bug-fix/SKILL.md +213 -0
  69. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-component-generator/SKILL.md +136 -0
  70. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-context-manager/SKILL.md +175 -0
  71. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-database-engineer/SKILL.md +104 -0
  72. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-deployment/SKILL.md +96 -0
  73. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-design-system/SKILL.md +137 -0
  74. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-form-builder/SKILL.md +140 -0
  75. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-frontend-architecture/SKILL.md +155 -0
  76. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-frontend-debug/SKILL.md +133 -0
  77. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-frontend-performance/SKILL.md +129 -0
  78. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-frontend-testing/SKILL.md +146 -0
  79. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-git-engineer/SKILL.md +304 -0
  80. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-help/SKILL.md +68 -0
  81. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-migration/SKILL.md +284 -0
  82. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-project-audit/SKILL.md +280 -0
  83. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-refactor/SKILL.md +222 -0
  84. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-state-management/SKILL.md +140 -0
  85. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-table-crud-generator/SKILL.md +127 -0
  86. package/.qk-ai-skill-os/skills/_archive_old_skills/qk-ui-builder/SKILL.md +152 -0
  87. package/.qk-ai-skill-os/skills/_template/SKILL.md +238 -0
  88. package/.qk-ai-skill-os/skills/qk-access-policy/SKILL.md +179 -0
  89. package/.qk-ai-skill-os/skills/qk-ai-builder/SKILL.md +221 -0
  90. package/.qk-ai-skill-os/skills/qk-api-lifecycle/SKILL.md +176 -0
  91. package/.qk-ai-skill-os/skills/qk-bug-resolution/SKILL.md +307 -0
  92. package/.qk-ai-skill-os/skills/qk-context-loader/SKILL.md +218 -0
  93. package/.qk-ai-skill-os/skills/qk-data-lifecycle/SKILL.md +192 -0
  94. package/.qk-ai-skill-os/skills/qk-db-optimizer/SKILL.md +196 -0
  95. package/.qk-ai-skill-os/skills/qk-design-to-code/SKILL.md +291 -0
  96. package/.qk-ai-skill-os/skills/qk-docs/SKILL.md +198 -0
  97. package/.qk-ai-skill-os/skills/qk-engineering-standard/SKILL.md +351 -0
  98. package/.qk-ai-skill-os/skills/qk-engineering-standard/rules/backend.md +122 -0
  99. package/.qk-ai-skill-os/skills/qk-engineering-standard/rules/database.md +3 -0
  100. package/.qk-ai-skill-os/skills/qk-engineering-standard/rules/frontend.md +152 -0
  101. package/.qk-ai-skill-os/skills/qk-engineering-standard/rules/security.md +3 -0
  102. package/.qk-ai-skill-os/skills/qk-engineering-standard/rules/testing.md +3 -0
  103. package/.qk-ai-skill-os/skills/qk-fe-api-integration/SKILL.md +326 -0
  104. package/.qk-ai-skill-os/skills/qk-feature-delivery/SKILL.md +305 -0
  105. package/.qk-ai-skill-os/skills/qk-help/SKILL.md +193 -0
  106. package/.qk-ai-skill-os/skills/qk-orchestrator/SKILL.md +277 -0
  107. package/.qk-ai-skill-os/skills/qk-orchestrator/references/routing-table.md +77 -0
  108. package/.qk-ai-skill-os/skills/qk-production-release/SKILL.md +290 -0
  109. package/.qk-ai-skill-os/skills/qk-project-bootstrap/SKILL.md +240 -0
  110. package/.qk-ai-skill-os/skills/qk-project-health/SKILL.md +199 -0
  111. package/.qk-ai-skill-os/skills/qk-project-memory/SKILL.md +172 -0
  112. package/.qk-ai-skill-os/skills/qk-system-evolution/SKILL.md +285 -0
  113. package/.qk-ai-skill-os/skills/qk-ui-audit/SKILL.md +314 -0
  114. package/.qk-ai-skill-os/skills/qk-ui-audit/references/anti-slop-checklist.md +136 -0
  115. package/.qk-ai-skill-os/skills/qk-ui-system-builder/SKILL.md +225 -0
  116. package/.qk-ai-skill-os/skills/qk-validation-gate/SKILL.md +361 -0
  117. package/.qk-ai-skill-os/skills.json +819 -0
  118. package/CLAUDE.md +110 -0
  119. package/README.md +25 -3
  120. package/add_lang.js +21 -0
  121. package/add_lang.py +25 -0
  122. package/add_sections.py +53 -0
  123. package/bin/install.js +225 -170
  124. package/bin/lint.js +122 -0
  125. package/docs/CHI_TIET_SKILLS.md +26 -26
  126. package/docs/HUONG_DAN_SU_DUNG.md +3 -3
  127. package/docs/SPEC.md +70 -20
  128. package/framework/skill-schema.md +310 -0
  129. package/package.json +15 -4
  130. package/patch.js +15 -0
  131. package/patch.py +89 -0
  132. package/patch2.py +83 -0
  133. package/skills/_archive_old_skills/qk-accessibility-audit/SKILL.md +1 -0
  134. package/skills/_archive_old_skills/qk-agent-orchestrator/SKILL.md +1 -0
  135. package/skills/_archive_old_skills/qk-api-integration/SKILL.md +1 -0
  136. package/skills/_archive_old_skills/qk-auth-security/SKILL.md +1 -0
  137. package/skills/_archive_old_skills/qk-backend-architecture/SKILL.md +1 -0
  138. package/skills/_archive_old_skills/qk-bug-fix/SKILL.md +1 -0
  139. package/skills/_archive_old_skills/qk-component-generator/SKILL.md +1 -0
  140. package/skills/_archive_old_skills/qk-context-manager/SKILL.md +1 -0
  141. package/skills/_archive_old_skills/qk-database-engineer/SKILL.md +1 -0
  142. package/skills/_archive_old_skills/qk-deployment/SKILL.md +1 -0
  143. package/skills/_archive_old_skills/qk-design-system/SKILL.md +1 -0
  144. package/skills/_archive_old_skills/qk-form-builder/SKILL.md +1 -0
  145. package/skills/_archive_old_skills/qk-frontend-architecture/SKILL.md +1 -0
  146. package/skills/_archive_old_skills/qk-frontend-debug/SKILL.md +1 -0
  147. package/skills/_archive_old_skills/qk-frontend-performance/SKILL.md +1 -0
  148. package/skills/_archive_old_skills/qk-frontend-testing/SKILL.md +1 -0
  149. package/skills/_archive_old_skills/qk-git-engineer/SKILL.md +1 -0
  150. package/skills/_archive_old_skills/qk-help/SKILL.md +1 -0
  151. package/skills/_archive_old_skills/qk-migration/SKILL.md +1 -0
  152. package/skills/_archive_old_skills/qk-project-audit/SKILL.md +1 -0
  153. package/skills/_archive_old_skills/qk-refactor/SKILL.md +1 -0
  154. package/skills/_archive_old_skills/qk-state-management/SKILL.md +1 -0
  155. package/skills/_archive_old_skills/qk-table-crud-generator/SKILL.md +1 -0
  156. package/skills/_archive_old_skills/qk-ui-builder/SKILL.md +1 -0
  157. package/skills/_template/SKILL.md +238 -0
  158. package/skills/qk-access-policy/SKILL.md +179 -39
  159. package/skills/qk-ai-builder/SKILL.md +215 -40
  160. package/skills/qk-api-lifecycle/SKILL.md +176 -46
  161. package/skills/qk-bug-resolution/SKILL.md +307 -46
  162. package/skills/qk-context-loader/SKILL.md +218 -43
  163. package/skills/qk-data-lifecycle/SKILL.md +192 -44
  164. package/skills/qk-db-optimizer/SKILL.md +196 -41
  165. package/skills/qk-design-to-code/SKILL.md +285 -46
  166. package/skills/qk-docs/SKILL.md +198 -40
  167. package/skills/qk-engineering-standard/SKILL.md +351 -42
  168. package/skills/qk-fe-api-integration/SKILL.md +326 -55
  169. package/skills/qk-feature-delivery/SKILL.md +305 -48
  170. package/skills/qk-help/SKILL.md +178 -23
  171. package/skills/qk-orchestrator/SKILL.md +277 -42
  172. package/skills/qk-orchestrator/references/routing-table.md +77 -0
  173. package/skills/qk-production-release/SKILL.md +284 -41
  174. package/skills/qk-project-bootstrap/SKILL.md +234 -38
  175. package/skills/qk-project-health/SKILL.md +199 -40
  176. package/skills/qk-project-memory/SKILL.md +172 -40
  177. package/skills/qk-system-evolution/SKILL.md +281 -40
  178. package/skills/qk-ui-audit/SKILL.md +314 -42
  179. package/skills/qk-ui-audit/references/anti-slop-checklist.md +136 -0
  180. package/skills/qk-ui-system-builder/SKILL.md +221 -43
  181. package/skills/qk-validation-gate/SKILL.md +359 -40
  182. package/skills.json +813 -824
  183. package/specs/fixtures/user-payload.json +12 -0
  184. package/specs/regressions/api-integration-null-fields.yaml +19 -0
  185. package/temp_fix.js +61 -0
  186. package/tests/registry.test.js +1 -1
  187. package/update_template.js +28 -0
  188. package/skills/qk-policy-engine/SKILL.md +0 -39
@@ -0,0 +1,326 @@
1
+ ---
2
+ name: qk-fe-api-integration
3
+ category: frontend
4
+ version: 7.5.0
5
+ description: "Consume API Backend, quản lý State, bind vào UI — tuân thủ kiến trúc Base dự án"
6
+ platforms: [antigravity, claude-code, cursor, windsurf, kilo-code]
7
+ execution_mode: deterministic
8
+
9
+ cost: medium
10
+ latency: medium
11
+ risk: low
12
+ side_effects: edit_files
13
+ produces: [code]
14
+ consumes: [json-payload, context-graph]
15
+
16
+ token_budget:
17
+ max_files_read: 3
18
+ max_lines_per_read: 100
19
+ max_shell_commands: 1
20
+ stop_early: true
21
+
22
+ exit_codes: [SUCCESS, BLOCKED, FAILED, PARTIAL]
23
+ skill_version: 7.5.0
24
+ runtime_version: 1
25
+ schema_version: 2
26
+ ---
27
+
28
+ # qk-fe-api-integration — Frontend API Consumer
29
+
30
+ > **Language rule:** Code, identifiers, file names → English. Explanations, summaries → Vietnamese.
31
+
32
+ ---
33
+
34
+ ## Preconditions
35
+ - [ ] Backend JSON payload hoặc API endpoint schema được cung cấp
36
+ - [ ] Project's API client pattern được xác định (Axios wrapper, RTK Query, React Query, etc.)
37
+
38
+ ```
39
+ On missing precondition:
40
+ EXIT: BLOCKED
41
+ Message: "Cần cung cấp: JSON response mẫu VÀ xác nhận API client pattern hiện tại của dự án."
42
+ ```
43
+
44
+ ---
45
+
46
+ ## Scope
47
+ - ✅ Identify existing API client pattern (NEVER default to raw fetch if wrapper exists)
48
+ - ✅ Generate strict TypeScript interfaces from JSON payload
49
+ - ✅ Implement service layer (API calls separated from UI)
50
+ - ✅ Handle all 3 UI states: Loading, Success, Error
51
+ - ✅ Map backend payload → DTO before injecting into UI components
52
+
53
+ ## Non-Goals
54
+ - ❌ Hardcode API Base URLs in UI components — use env vars or centralized config
55
+ - ❌ Inject API logic inside presentational (dumb) components
56
+ - ❌ Introduce new state management libraries without explicit user approval
57
+ - ❌ Hallucinate data fields not in provided JSON payload
58
+ - ❌ Create Node.js/Python scripts to patch — edit files directly
59
+ - ❌ Read entire files > 100 lines — use targeted reads
60
+ - ❌ Default to raw fetch if wrapper exists — ALWAYS reuse existing client
61
+
62
+ ---
63
+
64
+ ## Priority Order
65
+
66
+ | Priority | Task | Skip Threshold |
67
+ |----------|------|----------------|
68
+ | P1 | Identify existing API client (grep package.json, src/api/) | Never |
69
+ | P2 | Generate TypeScript interfaces from JSON payload | Never |
70
+ | P3 | Implement service layer (API calls) | Never |
71
+ | P4 | Bind to UI with Loading + Success + Error states | Budget < 30% |
72
+ | P5 | Cache invalidation after mutations (POST/PUT/DELETE) | Budget < 60% |
73
+ | P6 | DTO mapping (backend payload → clean UI data) | Budget < 70% |
74
+
75
+ ---
76
+
77
+ ## Workflow
78
+
79
+ ### Phase 1 — Identify Base Architecture
80
+
81
+ **Steps:**
82
+ 1. `grep_search` in `package.json` → identify: axios, react-query, rtk-query, swr, or custom
83
+ 2. `view_file[src/api/]` (first 30 lines) → confirm wrapper pattern
84
+ 3. Identify: base URL config, auth header injection, error interceptor location
85
+
86
+ **Decision:**
87
+ ```
88
+ IF wrapper/client found (e.g., apiClient.ts, axiosInstance.ts)
89
+ → Use it. NEVER bypass with raw fetch/axios.
90
+ → Confidence: HIGH → go to Phase 2
91
+
92
+ ELSE IF no wrapper found
93
+ → Confirm with user: "Dùng raw fetch hay tạo wrapper mới?"
94
+ → Confidence: MEDIUM → go to Phase 2
95
+
96
+ ELSE IF conflicting patterns found (mix of fetch + axios + rtk)
97
+ → EXIT: BLOCKED — ask which pattern to follow
98
+ ```
99
+
100
+ **Exit When:**
101
+ - API client pattern identified → go to Phase 2
102
+ - `max_files_read` reached → go to Phase 2 with MEDIUM confidence
103
+
104
+ ---
105
+
106
+ ### Phase 2 — Generate Types
107
+
108
+ **Steps:**
109
+ 1. Parse provided JSON payload → extract all fields with types
110
+ 2. Generate TypeScript interface (Request + Response)
111
+ 3. Flag any field that could be `null` or optional
112
+
113
+ **Decision:**
114
+ ```
115
+ IF all fields clearly typed from JSON
116
+ → Confidence: HIGH → go to Phase 3
117
+
118
+ ELSE IF some fields ambiguous (null | undefined)
119
+ → Mark as optional (?:) + add comment "// verify with backend"
120
+ → Confidence: MEDIUM → go to Phase 3
121
+
122
+ ELSE IF nested objects contain mixed null/non-null patterns
123
+ → EXIT: BLOCKED — cannot safely generate types without backend clarification
124
+ ```
125
+
126
+ ---
127
+
128
+ ### Phase 3 — Implement Service + UI Binding
129
+
130
+ **Steps:**
131
+ 1. Create/update service file (API calls only — no UI logic)
132
+ 2. Bind to component: implement Loading state (skeleton/spinner)
133
+ 3. Bind Success state: render data using generated types
134
+ 4. Bind Error state: handle HTTP errors per table below
135
+
136
+ **HTTP Error Handling (mandatory for ALL integrations):**
137
+ ```
138
+ 401 Unauthorized → redirect to login or show auth error toast
139
+ 403 Forbidden → show permission error toast
140
+ 404 Not Found → show empty state component (not a crash)
141
+ 422 Validation → map field errors to form fields
142
+ 500 Server Error → show generic error toast + retry button
143
+ Network Error → show offline banner or retry prompt
144
+ ```
145
+
146
+ **Decision:**
147
+ ```
148
+ IF all 3 states (Loading, Success, Error) implemented + service created → go to Phase 4
149
+ ELSE IF error state incomplete → EXIT: PARTIAL — missing HTTP error handling
150
+ ELSE → EXIT: BLOCKED — cannot complete binding
151
+ ```
152
+
153
+ **Exit When:**
154
+ - All 3 states implemented + service created → go to Phase 4
155
+
156
+ ---
157
+
158
+ ### Phase 4 — Self-Audit
159
+
160
+ **Steps:**
161
+ 1. Re-read service file: confirm no hardcoded URLs, no UI logic
162
+ 2. Re-read component: confirm no API calls inside JSX
163
+ 3. Verify cache invalidation if mutation (POST/PUT/DELETE)
164
+
165
+ **Decision:**
166
+ ```
167
+ IF all audit points pass → EXIT: SUCCESS
168
+ ELSE IF 1–2 minor issues → fix inline → EXIT: SUCCESS
169
+ ELSE IF architectural violation found → fix → EXIT: PARTIAL
170
+ ELSE IF hardcoded URL (http://localhost, 127.0.0.1) found → EXIT: FAILED — do not ship
171
+ ```
172
+
173
+ ---
174
+
175
+ ## Confidence Model
176
+
177
+ | Level | Condition | Action |
178
+ |-------|-----------|--------|
179
+ | HIGH | API client found, JSON payload complete, all types inferred | Implement directly |
180
+ | MEDIUM | Some fields ambiguous, wrapper partially identified | Note assumptions, proceed |
181
+ | LOW | No JSON payload provided OR conflicting API patterns OR nested nulls unresolved | EXIT: BLOCKED |
182
+
183
+ ---
184
+
185
+ ## Severity
186
+
187
+ | Level | Definition | Example |
188
+ |-------|-----------|---------|
189
+ | CRITICAL | Security data leakage from improper DTO mapping | Auth token in dumb component |
190
+ | HIGH | Missing error state causes white screen of death | No 500 handler |
191
+ | MEDIUM | Hardcoded URL breaks in production | `axios.get("http://localhost:3000/api")` |
192
+ | LOW | Missing cache invalidation after mutation | Stale data after POST |
193
+
194
+ ---
195
+
196
+ ## Evidence Format
197
+
198
+ ```
199
+ [SEVERITY] path/to/component.tsx:LINE
200
+ Issue: [specific violation]
201
+ Confidence: [HIGH|MEDIUM|LOW]
202
+ Fix: [one-line suggestion]
203
+ ```
204
+
205
+ **Example:**
206
+ ```
207
+ [HIGH] src/pages/Dashboard.tsx:45
208
+ Issue: API call `axios.get('/api/data')` inside JSX render — violates smart/dumb separation
209
+ Confidence: HIGH
210
+ Fix: Move to `src/services/dashboard.service.ts`, inject via hook
211
+ ```
212
+
213
+ ---
214
+
215
+ ## Retry Policy
216
+
217
+ ```
218
+ API call fails (runtime verification)
219
+ └─ Check error type
220
+ ├─ Network error → show retry button (user-triggered retry, not auto)
221
+ ├─ 401/403 → do NOT retry — redirect or show permission error
222
+ ├─ 500 → auto-retry once after 1s delay
223
+ │ └─ Still fails → show error toast with "Try again" button
224
+ └─ Never auto-retry more than 1 time (avoid hammering server)
225
+
226
+ Skill execution fails
227
+ └─ Check failure type
228
+ ├─ Syntax error in generated code → fix and retry once
229
+ └─ Architectural mismatch → EXIT: FAILED, do NOT retry
230
+ ```
231
+
232
+ ---
233
+
234
+ ## Escalation Rules
235
+
236
+ ```
237
+ BLOCKED: Cannot determine API client pattern
238
+ Missing:
239
+ - package.json contents OR
240
+ - Location of existing API client file (e.g., src/lib/api.ts)
241
+ Questions:
242
+ 1. Dự án dùng thư viện nào để gọi API? (axios / fetch / react-query / rtk-query)
243
+ 2. Có file API wrapper/client nào sẵn không? (ví dụ: src/api/client.ts)
244
+ Recommended Assumptions (if proceeding):
245
+ - Create new Axios instance with interceptors for auth + error handling
246
+ ```
247
+
248
+ ---
249
+
250
+ ## Handoff Contract
251
+
252
+ ### Consumes
253
+ ```json
254
+ {
255
+ "from": "user or qk-context-loader",
256
+ "required_fields": ["json_payload_sample", "api_endpoint"],
257
+ "optional_fields": ["existing_api_client_path", "component_to_bind"]
258
+ }
259
+ ```
260
+
261
+ ### Produces
262
+ ```json
263
+ {
264
+ "to": "user",
265
+ "output_fields": ["types_file", "service_file", "component_file", "states_implemented", "exit_code"]
266
+ }
267
+ ```
268
+
269
+ ---
270
+
271
+ ## Output Format
272
+
273
+ ```
274
+ 🔌 FE API Integration
275
+ ─────────────────────────────────────────────────
276
+ API Client: [identified — axiosInstance.ts | created new]
277
+ Types: [src/types/[resource].types.ts]
278
+ Service: [src/services/[resource].service.ts]
279
+
280
+ States implemented:
281
+ ✅ Loading (skeleton/spinner)
282
+ ✅ Success (data rendered)
283
+ ✅ Error (toast/empty state per HTTP code)
284
+ ⚠️ [any state skipped — reason]
285
+
286
+ Assumptions:
287
+ - [any assumption made]
288
+
289
+ Exit Code: [SUCCESS | PARTIAL | BLOCKED | FAILED]
290
+ ```
291
+
292
+ ---
293
+
294
+ ## Exit Codes
295
+
296
+ | Code | Meaning | When |
297
+ |------|---------|------|
298
+ | SUCCESS | Types + service + UI binding complete with all 3 states | Full integration |
299
+ | PARTIAL | Integration done, missing error/loading state or cache invalidation | Follow-up needed |
300
+ | BLOCKED | No JSON payload or conflicting API patterns | Ask user |
301
+ | FAILED | Cannot integrate without breaking existing architecture | Escalate |
302
+
303
+ ---
304
+
305
+ Consume backend API safely in frontend: identify existing client, generate types, implement service layer, and bind to UI with proper states.
306
+ Triggered when user needs to integrate a backend API endpoint into a frontend application. Requires JSON payload sample and knowledge of existing API client patterns.
307
+ - JSON payload sample (request/response)
308
+ - API endpoint specification
309
+ - Existing API client path (if any)
310
+ - Component to bind (if specified)
311
+ - Context graph (for existing patterns)
312
+ 1. **Identify:** Find existing API client or confirm need for new one
313
+ 2. **Generate:** Create TypeScript interfaces from JSON payload
314
+ 3. **Implement:** Build service layer with proper HTTP methods
315
+ 4. **Bind:** Connect to UI with Loading/Success/Error states
316
+ 5. **Audit:** Verify no hardcoded URLs, no API in JSX, proper error handling
317
+ - NEVER bypass existing wrapper with raw fetch/axios
318
+ - MUST generate strict TypeScript interfaces (no `any`)
319
+ - MUST handle all 3 UI states: Loading, Success, Error
320
+ - MUST NOT exceed token_budget (max 3 files, 100 lines each, 1 shell command)
321
+ - MUST stop early if confidence threshold reached
322
+ - Zero-Trust: Use existing client pattern, never invent new one without approval
323
+ - Type Safety: All API responses must have explicit TypeScript types
324
+ - Error Handling: All HTTP errors must be handled per mandatory table
325
+ - Separation: Service layer only — no API calls in presentational components
326
+ ---
@@ -0,0 +1,305 @@
1
+ ---
2
+ name: qk-feature-delivery
3
+ category: fullstack
4
+ version: 7.5.0
5
+ description: "Phát triển tính năng mới end-to-end: Requirements → Context → Design → Implement → Self-audit."
6
+ platforms: [antigravity, claude-code, cursor, windsurf, kilo-code]
7
+ execution_mode: deterministic
8
+
9
+ cost: high
10
+ latency: slow
11
+ risk: medium
12
+ side_effects: edit_files
13
+ produces: [code, report]
14
+ consumes: [context-graph, design-md, user-description]
15
+
16
+ token_budget:
17
+ max_files_read: 5
18
+ max_lines_per_read: 150
19
+ max_shell_commands: 2
20
+ stop_early: false
21
+
22
+ exit_codes: [SUCCESS, BLOCKED, FAILED, PARTIAL]
23
+ skill_version: 7.5.0
24
+ runtime_version: 1
25
+ schema_version: 2
26
+ ---
27
+
28
+ # qk-feature-delivery — End-to-End Feature Builder
29
+
30
+ > **Language rule:** Code, identifiers, file names → English. Explanations, summaries → Vietnamese.
31
+
32
+ ---
33
+
34
+ ## Preconditions
35
+ - [ ] Feature requirement is specific enough to implement (not just "add something")
36
+ - [ ] `DESIGN.md` exists with required tokens if UI work is involved (colors, spacing, typography)
37
+ - [ ] Context graph available OR `qk-context-loader` run first (for logic work)
38
+
39
+ ```
40
+ On missing precondition:
41
+ EXIT: BLOCKED
42
+ Message: "[Precondition] missing. See Escalation Rules."
43
+ ```
44
+
45
+ ---
46
+
47
+ ## Scope
48
+ - ✅ Deliver a complete, verifiable feature end-to-end (FE + BE + API)
49
+ - ✅ Adhere to project architecture from context graph
50
+ - ✅ Adhere to DESIGN.md for any UI components
51
+ - ✅ Self-audit before marking done
52
+
53
+ ## Non-Goals
54
+ - ❌ Mix refactoring into feature work
55
+ - ❌ Add dependencies without asking user first
56
+ - ❌ Emit generic AI-slop UI or unverified logic
57
+ - ❌ Ignore backward compatibility on existing APIs
58
+ - ❌ Proceed with ambiguous requirements without user confirmation
59
+ - ❌ Break existing API contracts — version or deprecate instead
60
+
61
+ ---
62
+
63
+ ## Priority Order
64
+
65
+ | Priority | Task | Skip Threshold |
66
+ |----------|------|----------------|
67
+ | P1 | Core feature logic (happy path) | Never |
68
+ | P2 | Error/loading/empty UI states | Budget < 30% |
69
+ | P3 | Edge cases and validation | Budget < 50% |
70
+ | P4 | Regression prevention (existing tests) | Budget < 60% |
71
+ | P5 | Backend contract validation (API schema, DB migrations) | Budget < 60% |
72
+ | P6 | Documentation + comments | Budget < 70% |
73
+
74
+ ---
75
+
76
+ ## Workflow
77
+
78
+ ### Phase 1 — Requirements Clarification
79
+
80
+ **Steps:**
81
+ 1. Parse user request → extract: what, where, acceptance criteria
82
+ 2. Check ambiguity: Can I implement this without assumptions?
83
+
84
+ **Decision:**
85
+ ```
86
+ IF requirements are specific (what + where + expected behavior)
87
+ → Confidence: HIGH → go to Phase 2
88
+
89
+ ELSE IF 1–2 ambiguous points
90
+ → ASK USER clarifying questions
91
+ → Confidence: MEDIUM → WAIT for user response
92
+ → Do NOT proceed to Phase 2 until clarified
93
+
94
+ ELSE IF core requirement unclear
95
+ → EXIT: BLOCKED — ask clarifying questions
96
+ ```
97
+
98
+ **Exit When:**
99
+ - Requirements understood → go to Phase 2
100
+
101
+ ---
102
+
103
+ ### Phase 2 — Context Loading
104
+
105
+ **Steps:**
106
+ 1. If logic work: read context graph from `qk-context-loader` output
107
+ 2. Identify: entry point, affected modules, existing patterns to reuse
108
+ 3. If UI work: read `DESIGN.md` → extract tokens for this component
109
+ 4. Document rollback plan — which files to revert if feature fails
110
+
111
+ **Decision:**
112
+ ```
113
+ IF context graph available
114
+ → Use it, proceed to Phase 3
115
+
116
+ ELSE IF no context graph but simple task
117
+ → Simple task = modifies ≤ 1 file, no new dependencies, no new API contracts
118
+ → grep_search for entry point → minimal manual map
119
+ → Confidence: MEDIUM → proceed to Phase 3
120
+
121
+ ELSE IF complex task without context graph
122
+ → EXIT: BLOCKED — run qk-context-loader first
123
+ ```
124
+
125
+ ---
126
+
127
+ ### Phase 3 — Implementation
128
+
129
+ **Steps (in order):**
130
+ 1. Create/modify files using `replace_file_content` / `multi_replace_file_content`
131
+ 2. Follow existing code patterns (from context graph)
132
+ 3. For UI: use DESIGN.md tokens only — no hardcoded colors/sizes
133
+ 4. Implement backend error handling: try/catch, transaction rollback, input validation
134
+ 5. Handle all 3 UI states: Loading + Success + Error
135
+
136
+ **Exit When:**
137
+ - Happy path implemented → go to Phase 4
138
+ - Blocked by missing dependency → EXIT: BLOCKED, ask user
139
+
140
+ ---
141
+
142
+ ### Phase 4 — Self-Audit
143
+
144
+ **Steps:**
145
+ 1. Re-read the code written (targeted view, ≤ 150 lines)
146
+ 2. Check against: backward compat, DESIGN.md compliance, error states
147
+ 3. Verify test coverage meets project threshold (default: ≥ 80%)
148
+
149
+ **Decision:**
150
+ ```
151
+ IF all checks pass
152
+ → EXIT: SUCCESS
153
+
154
+ ELSE IF minor issues found (missing error state, style token)
155
+ → Fix immediately, EXIT: SUCCESS
156
+
157
+ ELSE IF significant issue (breaks existing API, wrong architecture)
158
+ → Fix, note change in output, EXIT: FAILED — do not ship breaking changes as PARTIAL
159
+
160
+ ELSE IF missing backend error handling
161
+ → Fix immediately, EXIT: SUCCESS
162
+ ```
163
+
164
+ ---
165
+
166
+ ## Confidence Model
167
+
168
+ | Level | Condition | Action |
169
+ |-------|-----------|--------|
170
+ | HIGH | Context graph available, DESIGN.md read, requirements clear | Implement directly |
171
+ | MEDIUM | Some assumptions made (≤ 2) | Note assumptions in output |
172
+ | LOW | Unclear requirements or missing context OR assumptions > 2 | EXIT: BLOCKED |
173
+
174
+ ---
175
+
176
+ ## Severity (for self-audit findings)
177
+
178
+ | Level | Definition |
179
+ |-------|-----------|
180
+ | CRITICAL | Feature breaks existing functionality |
181
+ | HIGH | Missing required states (error/loading) or missing backend error handling |
182
+ | MEDIUM | Not following DESIGN.md or project patterns |
183
+ | LOW | Missing comments or minor style issue |
184
+
185
+ ---
186
+
187
+ ## Evidence Format (Self-Audit)
188
+
189
+ ```
190
+ [SEVERITY] path/to/file.ts:LINE
191
+ Issue: [what was found]
192
+ Confidence: HIGH
193
+ Fix: [applied immediately OR noted for follow-up]
194
+ ```
195
+
196
+ ---
197
+
198
+ ## Escalation Rules
199
+
200
+ ```
201
+ BLOCKED: [specific reason]
202
+ Missing:
203
+ - [What's needed — context graph / DESIGN.md / clear requirements]
204
+ Questions:
205
+ 1. [Specific clarifying question]
206
+ 2. [Second question if needed]
207
+ Recommended Assumptions (if proceeding):
208
+ - [Safe assumption 1]
209
+ - [Safe assumption 2 — note these will be in output]
210
+ ```
211
+
212
+ ---
213
+
214
+ ## Handoff Contract
215
+
216
+ ### Consumes
217
+ ```json
218
+ {
219
+ "from": "qk-context-loader + user",
220
+ "required_fields": ["feature_description", "context_graph"],
221
+ "optional_fields": ["design_md_path", "acceptance_criteria"]
222
+ }
223
+ ```
224
+
225
+ ### Produces
226
+ ```json
227
+ {
228
+ "to": "qk-validation-gate",
229
+ "output_fields": ["changed_files", "assumptions_made", "self_audit_result", "exit_code"]
230
+ }
231
+ ```
232
+
233
+ ---
234
+
235
+ ## Output Format
236
+
237
+ ```
238
+ 🚀 Feature Delivery
239
+ ─────────────────────────────────────────────────
240
+ Feature: [feature name]
241
+ Confidence: [HIGH | MEDIUM]
242
+
243
+ Implementation:
244
+ ✅ [file:line — what was added/changed]
245
+ ✅ [file:line — what was added/changed]
246
+
247
+ Assumptions made:
248
+ - [assumption 1 — safe/risky]
249
+
250
+ Self-Audit:
251
+ ✅ Backward compatible
252
+ ✅ Error/Loading states handled
253
+ ✅ DESIGN.md tokens used
254
+ ✅ Backend error handling implemented
255
+ ⚠️ [Any noted issue]
256
+
257
+ Exit Code: [SUCCESS | PARTIAL | BLOCKED | FAILED]
258
+ ```
259
+
260
+ ---
261
+
262
+ ## Exit Codes
263
+
264
+ | Code | Meaning | When |
265
+ |------|---------|------|
266
+ | SUCCESS | Feature complete, self-audit passed | All states handled, patterns followed |
267
+ | PARTIAL | Feature works but with minor gaps | Medium confidence, some skipped |
268
+ | BLOCKED | Cannot proceed — missing context/design/requirements | Stop and ask user |
269
+ | FAILED | Cannot implement — breaks existing system | Architectural conflict or API break |
270
+
271
+ ---
272
+
273
+ Deliver a complete, verifiable feature end-to-end (FE + BE + API) while adhering to project architecture and DESIGN.md.
274
+ This skill is the primary execution skill for building new features. It requires clear requirements, context graph for architecture understanding, and DESIGN.md for UI work.
275
+ - Feature description with acceptance criteria
276
+ - Context graph (from qk-context-loader)
277
+ - DESIGN.md path (if UI involved)
278
+ - Existing codebase patterns
279
+ - User approval for new dependencies
280
+ 1. **Clarify:** Parse requirements, identify ambiguities, ask user if needed
281
+ 2. **Load:** Read context graph and DESIGN.md tokens
282
+ 3. **Implement:** Build feature following existing patterns
283
+ 4. **Audit:** Self-audit for backward compat, error states, DESIGN.md compliance
284
+ - MUST have context graph OR explicit confirmation for simple tasks
285
+ - MUST have DESIGN.md tokens if UI is involved
286
+ - MUST handle all 3 UI states: Loading, Success, Error
287
+ - MUST NOT exceed token_budget (max 5 files, 150 lines each, 2 shell commands)
288
+ - MUST NOT break existing API contracts without versioning
289
+ - Zero-Trust: No assumptions without user confirmation
290
+ - Backward Compat: Never break existing public APIs
291
+ - Design First: UI must use DESIGN.md tokens only
292
+ - Error Handling: Backend must have try/catch, validation, rollback
293
+ ## Retry Policy
294
+ ```
295
+ Fix applied
296
+ └─ Static verification (re-read fixed section)
297
+ ├─ PASS → EXIT: SUCCESS
298
+ └─ Issue detected → attempt 1 correction
299
+ └─ Re-verify
300
+ ├─ PASS → EXIT: SUCCESS
301
+ └─ FAIL → EXIT: PARTIAL + report both attempts
302
+ └─ Do NOT attempt 3rd fix — ESCALATE to user
303
+ ```
304
+
305
+ ---