ai-flow-dev 2.2.0 → 2.2.1
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.
- package/README.md +13 -17
- package/dist/cli.js +8 -2
- package/dist/cli.js.map +1 -1
- package/package.json +1 -1
- package/prompts/backend/flow-build-phase-0.md +286 -4
- package/prompts/backend/flow-build-phase-1.md +19 -0
- package/prompts/backend/flow-build-phase-2.md +19 -0
- package/prompts/backend/flow-build-phase-3.md +19 -0
- package/prompts/backend/flow-build-phase-4.md +19 -0
- package/prompts/backend/flow-build-phase-5.md +19 -0
- package/prompts/backend/flow-build-phase-6.md +19 -0
- package/prompts/backend/flow-build-phase-7.md +19 -0
- package/prompts/backend/flow-build-phase-9.md +14 -0
- package/prompts/backend/flow-check-review.md +20 -0
- package/prompts/backend/flow-check-test.md +14 -0
- package/prompts/backend/flow-check.md +65 -0
- package/prompts/backend/flow-commit.md +51 -0
- package/prompts/backend/flow-docs-sync.md +53 -53
- package/prompts/backend/flow-work-feature.md +42 -0
- package/prompts/backend/flow-work-fix.md +33 -0
- package/prompts/backend/flow-work-refactor.md +32 -0
- package/prompts/backend/flow-work-resume.md +32 -0
- package/prompts/backend/flow-work.md +127 -0
- package/prompts/frontend/flow-build-phase-0.md +363 -35
- package/prompts/frontend/flow-build-phase-1.md +433 -404
- package/prompts/frontend/flow-build-phase-2.md +508 -872
- package/prompts/frontend/flow-build-phase-3.md +629 -562
- package/prompts/frontend/flow-build-phase-4.md +438 -382
- package/prompts/frontend/flow-build-phase-5.md +559 -362
- package/prompts/frontend/flow-build-phase-6.md +383 -452
- package/prompts/frontend/flow-build-phase-7.md +818 -392
- package/prompts/frontend/flow-build-phase-9.md +14 -0
- package/prompts/frontend/flow-check-review.md +20 -0
- package/prompts/frontend/flow-check-test.md +14 -0
- package/prompts/frontend/flow-check.md +65 -0
- package/prompts/frontend/flow-commit.md +51 -0
- package/prompts/frontend/flow-docs-sync.md +36 -34
- package/prompts/frontend/flow-work-feature.md +42 -0
- package/prompts/frontend/flow-work-fix.md +33 -0
- package/prompts/frontend/flow-work-refactor.md +32 -0
- package/prompts/frontend/flow-work-resume.md +32 -0
- package/prompts/frontend/flow-work.md +127 -0
- package/prompts/mobile/flow-build-phase-0.md +366 -37
- package/prompts/mobile/flow-build-phase-1.md +438 -493
- package/prompts/mobile/flow-build-phase-2.md +458 -464
- package/prompts/mobile/flow-build-phase-3.md +613 -487
- package/prompts/mobile/flow-build-phase-4.md +439 -258
- package/prompts/mobile/flow-build-phase-5.md +582 -250
- package/prompts/mobile/flow-build-phase-6.md +389 -359
- package/prompts/mobile/flow-build-phase-7.md +871 -285
- package/prompts/mobile/flow-build-phase-9.md +14 -0
- package/prompts/mobile/flow-check-review.md +20 -0
- package/prompts/mobile/flow-check-test.md +14 -0
- package/prompts/mobile/flow-check.md +65 -0
- package/prompts/mobile/flow-commit.md +51 -0
- package/prompts/mobile/flow-docs-sync.md +37 -40
- package/prompts/mobile/flow-work-feature.md +42 -0
- package/prompts/mobile/flow-work-fix.md +33 -0
- package/prompts/mobile/flow-work-refactor.md +32 -0
- package/prompts/mobile/flow-work-resume.md +32 -0
- package/prompts/mobile/flow-work.md +127 -0
- package/prompts/shared/smart-skip-preflight.md +214 -0
- package/prompts/backend/flow-dev-commit.md +0 -829
- package/prompts/backend/flow-dev-feature.md +0 -1948
- package/prompts/backend/flow-dev-fix.md +0 -952
- package/prompts/backend/flow-dev-refactor.md +0 -690
- package/prompts/backend/flow-dev-review.md +0 -372
- package/prompts/backend/flow-dev-work.md +0 -1081
|
@@ -73,6 +73,20 @@ Generate atomic tasks using the `{{VARIABLE}}` paths.
|
|
|
73
73
|
### Step 9.5: Validate Coverage
|
|
74
74
|
Check for gaps.
|
|
75
75
|
|
|
76
|
+
## Architectural Markers Key
|
|
77
|
+
|
|
78
|
+
| Marker | Name | Description |
|
|
79
|
+
|--------|------|-------------|
|
|
80
|
+
| **[U]** | UI/Comp | Atoms, Molecules, Organisms |
|
|
81
|
+
| **[P]** | Page | Page layout, route implementation |
|
|
82
|
+
| **[L]** | Logic/Hook | Custom hooks, business logic |
|
|
83
|
+
| **[S]** | State | Store actions, reducers, context |
|
|
84
|
+
| **[A]** | API/Data | Fetching, services, transformers |
|
|
85
|
+
| **[T]** | Test | Unit, component, e2e tests |
|
|
86
|
+
| **[D]** | Docs | Documentation updates |
|
|
87
|
+
|
|
88
|
+
---
|
|
89
|
+
|
|
76
90
|
### Step 9.6: Generate Document (docs/roadmap.md)
|
|
77
91
|
|
|
78
92
|
---
|
|
@@ -0,0 +1,20 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Internal logic for Code Review within flow-check
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Review Logic
|
|
6
|
+
|
|
7
|
+
Performs professional multi-aspect code review.
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
## 🔍 Review Aspects
|
|
11
|
+
1. **Security**: OWASP patterns, secrets, auth.
|
|
12
|
+
2. **Performance**: N+1, leaks, async blocking.
|
|
13
|
+
3. **Tests**: Edge cases, quality, coverage.
|
|
14
|
+
4. **Architecture**: SOLID, DRY, Patterns.
|
|
15
|
+
5. **Quality**: Naming, complexity, linting.
|
|
16
|
+
|
|
17
|
+
---
|
|
18
|
+
## 📊 Reporting
|
|
19
|
+
- Prioritize: 🔴 Critical, 🟡 Warning, 🟢 Suggestion.
|
|
20
|
+
- Save report to `.ai-flow/reviews/`.
|
|
@@ -0,0 +1,14 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Internal logic for Testing within flow-check
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Testing Logic
|
|
6
|
+
|
|
7
|
+
Executes automated tests and analyzes results.
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
## 🧪 Test Flow
|
|
11
|
+
1. Identify project test runner (detected from `package.json`, etc.).
|
|
12
|
+
2. Execute full suite or targeted tests.
|
|
13
|
+
3. Parse results (Passing/Failing/Coverage).
|
|
14
|
+
4. Update `status.json` validation section.
|
|
@@ -0,0 +1,65 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Combined Validation workflow for Tests and Code Review
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Unified Check Workflow
|
|
6
|
+
|
|
7
|
+
**YOU ARE AN EXPERT QUALITY ASSURANCE AND CODE REVIEW SPECIALIST.**
|
|
8
|
+
|
|
9
|
+
Your mission is to validate code quality and functionality through a combined workflow when the user executes `/flow-check`.
|
|
10
|
+
|
|
11
|
+
---
|
|
12
|
+
## Command: `/flow-check`
|
|
13
|
+
|
|
14
|
+
### Objective
|
|
15
|
+
Provide a comprehensive validation suite including automated tests and professional code review in a single, prioritized report.
|
|
16
|
+
|
|
17
|
+
---
|
|
18
|
+
## Workflow: 3 Stages
|
|
19
|
+
|
|
20
|
+
### Stage 1: Automated Testing & Analysis
|
|
21
|
+
- **Tests**: Execute `npm test` (or project equivalent).
|
|
22
|
+
- **Lint**: Execute linting rules.
|
|
23
|
+
- **Types**: Execute type checking (`tsc`).
|
|
24
|
+
- **Update `status.json`**: Record results in `validation` section.
|
|
25
|
+
|
|
26
|
+
### Stage 2: Professional Code Review
|
|
27
|
+
Analyze changes from 5 perspectives:
|
|
28
|
+
1. **🔒 Security**: SQLi, XSS, Secrets, Auth patterns.
|
|
29
|
+
2. **⚡ Performance**: N+1 queries, memory leaks, blocking ops.
|
|
30
|
+
3. **🧪 Testing**: Coverage, edge cases, test quality.
|
|
31
|
+
4. **📐 Architecture**: SOLID, DRY, Coupling.
|
|
32
|
+
5. **🎨 Code Quality**: Naming, complexity, consistency.
|
|
33
|
+
|
|
34
|
+
### Stage 3: Summary Report
|
|
35
|
+
Generate a prioritized report:
|
|
36
|
+
- 🔴 **Critical Issues**: Fix immediately.
|
|
37
|
+
- 🟡 **Warnings**: Fix before merge.
|
|
38
|
+
- 🟢 **Suggestions**: Improvement opportunities.
|
|
39
|
+
|
|
40
|
+
---
|
|
41
|
+
## Integration with `status.json`
|
|
42
|
+
|
|
43
|
+
Automatically update the `validation` section:
|
|
44
|
+
```json
|
|
45
|
+
"validation": {
|
|
46
|
+
"tests": { "executed": true, "passed": true, "summary": "X/Y passed" },
|
|
47
|
+
"lint": { "executed": true, "passed": true },
|
|
48
|
+
"typeCheck": { "executed": true, "passed": true }
|
|
49
|
+
}
|
|
50
|
+
```
|
|
51
|
+
|
|
52
|
+
And update the `finalChecklist`:
|
|
53
|
+
```json
|
|
54
|
+
"finalChecklist": {
|
|
55
|
+
"testsComplete": true
|
|
56
|
+
}
|
|
57
|
+
```
|
|
58
|
+
|
|
59
|
+
---
|
|
60
|
+
## Delegation
|
|
61
|
+
- Detailed Test execution → `@flow-check-test.md`
|
|
62
|
+
- Detailed Review logic → `@flow-check-review.md`
|
|
63
|
+
|
|
64
|
+
---
|
|
65
|
+
**BEGIN EXECUTION when user runs `/flow-check`**
|
|
@@ -0,0 +1,51 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Conventional Commits Automation
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Commit Automation
|
|
6
|
+
|
|
7
|
+
**YOU ARE AN EXPERT GIT WORKFLOW SPECIALIST.**
|
|
8
|
+
|
|
9
|
+
Your mission is to analyze changes, group them intelligently, and create atomic commits following Conventional Commits standard when the user executes `/flow-commit`.
|
|
10
|
+
|
|
11
|
+
---
|
|
12
|
+
## Command: `/flow-commit`
|
|
13
|
+
|
|
14
|
+
### Objective
|
|
15
|
+
Automate commit creation with:
|
|
16
|
+
- **Intelligent grouping** by functional relationship.
|
|
17
|
+
- **Conventional Commits** compliance (type, scope, description).
|
|
18
|
+
- **Atomic commits** (one logical change per commit).
|
|
19
|
+
|
|
20
|
+
---
|
|
21
|
+
## Workflow: 4 Steps
|
|
22
|
+
|
|
23
|
+
### Step 1: Detect Changes
|
|
24
|
+
- Modified files (unstaged/staged).
|
|
25
|
+
- **Untracked files** (`git status --porcelain`).
|
|
26
|
+
- Deleted files.
|
|
27
|
+
|
|
28
|
+
### Step 2: Intelligent Grouping
|
|
29
|
+
Group files by:
|
|
30
|
+
- **Feature Complete**: Entity + Service + Controller + Tests + Docs.
|
|
31
|
+
- **Refactoring**: Helper + Tests + Usages.
|
|
32
|
+
- **Configuration**: Docker, Env, CI/CD.
|
|
33
|
+
- **Independent Tests/Docs**.
|
|
34
|
+
|
|
35
|
+
### Step 3: Create Commits
|
|
36
|
+
1. Generate Conventional Commit message.
|
|
37
|
+
2. `git add [files] && git commit -m "[message]"`.
|
|
38
|
+
3. **Wait for user confirmation.**
|
|
39
|
+
|
|
40
|
+
### Step 4: Summary & Push
|
|
41
|
+
- Show `git log` of new commits.
|
|
42
|
+
- Suggest `git push origin [branch]`.
|
|
43
|
+
|
|
44
|
+
---
|
|
45
|
+
## Integration with `status.json`
|
|
46
|
+
- Update `git.commits` array.
|
|
47
|
+
- Set `git.uncommittedChanges: false` when done.
|
|
48
|
+
- Update `finalChecklist.committed: true`.
|
|
49
|
+
|
|
50
|
+
---
|
|
51
|
+
**BEGIN EXECUTION when user runs `/flow-commit`**
|
|
@@ -18,29 +18,33 @@ Detect changes in the frontend codebase compared to the last documented state (s
|
|
|
18
18
|
|
|
19
19
|
### Step 1: Check for Analysis File
|
|
20
20
|
|
|
21
|
+
// turbo
|
|
22
|
+
```bash
|
|
23
|
+
cat .ai-flow/cache/docs-analysis.json
|
|
21
24
|
```
|
|
22
|
-
First, check if `.ai-flow/cache/docs-analysis.json` exists:
|
|
23
25
|
|
|
24
26
|
- ✅ If exists → Proceed to Step 2 (Compare Changes)
|
|
25
27
|
- ❌ If NOT exists → Execute full Phase 0 analysis first:
|
|
26
|
-
- Run complete frontend code analysis (
|
|
28
|
+
- Run complete frontend code analysis (Project Discovery)
|
|
27
29
|
- Create `.ai-flow/cache/docs-analysis.json` with current state
|
|
28
30
|
- Then proceed to Step 2
|
|
29
|
-
```
|
|
30
31
|
|
|
31
32
|
### Step 2: Detect Changes
|
|
32
33
|
|
|
33
34
|
**Reuse Phase 0 Analysis Logic:**
|
|
34
35
|
|
|
35
36
|
1. **Perform Current Code Analysis:**
|
|
36
|
-
|
|
37
|
-
|
|
38
|
-
|
|
39
|
-
-
|
|
40
|
-
|
|
41
|
-
|
|
42
|
-
-
|
|
43
|
-
-
|
|
37
|
+
- Execute project-wide discovery using cross-platform commands:
|
|
38
|
+
// turbo
|
|
39
|
+
```bash
|
|
40
|
+
ls -R . --exclude-standard
|
|
41
|
+
```
|
|
42
|
+
- Analyze current state for:
|
|
43
|
+
- View & UI components architecture
|
|
44
|
+
- Interface definitions (Routing patterns, public exports, entry points)
|
|
45
|
+
- State management and data flow (Global stores, Contexts, Hooks)
|
|
46
|
+
- Style definitions and design tokens
|
|
47
|
+
- Core dependencies and configuration
|
|
44
48
|
- Generate current state snapshot
|
|
45
49
|
|
|
46
50
|
2. **Compare with Previous State:**
|
|
@@ -48,12 +52,12 @@ First, check if `.ai-flow/cache/docs-analysis.json` exists:
|
|
|
48
52
|
- Load `.ai-flow/cache/docs-analysis.json`
|
|
49
53
|
- Compare current state vs previous state
|
|
50
54
|
- Detect changes in:
|
|
51
|
-
- **
|
|
52
|
-
- **
|
|
53
|
-
- **
|
|
54
|
-
- **
|
|
55
|
-
- **
|
|
56
|
-
- **Configuration:** New environment
|
|
55
|
+
- **Interfaces:** New, modified, or deleted "entry points" (Routes, Exports)
|
|
56
|
+
- **UI Structures:** New, modified, or deleted views/components
|
|
57
|
+
- **State Logic:** New, modified, or deleted state containers/logic
|
|
58
|
+
- **Style Forms:** New, modified, or deleted styling tokens/approaches
|
|
59
|
+
- **Dependencies:** Manifest changes (version bumps, new packages)
|
|
60
|
+
- **Configuration:** New environment keys or external integrations
|
|
57
61
|
|
|
58
62
|
3. **Generate Change Report:**
|
|
59
63
|
- Categorize changes by type
|
|
@@ -189,20 +193,19 @@ Update cancelled. Run `/flow-docs-sync` when you're ready to update the document
|
|
|
189
193
|
---
|
|
190
194
|
## Change Detection Rules
|
|
191
195
|
|
|
192
|
-
###
|
|
196
|
+
### UI & View Detection (Agnostic)
|
|
193
197
|
|
|
194
|
-
**What triggers `docs/components.md`
|
|
198
|
+
**What triggers document update (e.g., `docs/components.md`):**
|
|
195
199
|
|
|
196
|
-
- New component
|
|
197
|
-
-
|
|
198
|
-
- Modified component exports or patterns
|
|
200
|
+
- New component definitions or file markers (e.g., standard exports, folder structures)
|
|
201
|
+
- Modified component hierarchies or patterns
|
|
199
202
|
- Deleted components
|
|
200
203
|
|
|
201
204
|
**How to update:**
|
|
202
205
|
|
|
203
|
-
-
|
|
204
|
-
-
|
|
205
|
-
- Maintain
|
|
206
|
+
- Update component catalogs following established project conventions
|
|
207
|
+
- Refresh hierarchy diagrams (Mermaid) to reflect new structures
|
|
208
|
+
- Maintain existing documentation for stable components
|
|
206
209
|
|
|
207
210
|
### State Management Detection
|
|
208
211
|
|
|
@@ -220,20 +223,19 @@ Update cancelled. Run `/flow-docs-sync` when you're ready to update the document
|
|
|
220
223
|
- Update state patterns if changed
|
|
221
224
|
- Maintain all existing state management documentation
|
|
222
225
|
|
|
223
|
-
###
|
|
226
|
+
### Style & Design Detection (Agnostic)
|
|
224
227
|
|
|
225
|
-
**What triggers `docs/styling.md`
|
|
228
|
+
**What triggers document update (e.g., `docs/styling.md`):**
|
|
226
229
|
|
|
227
|
-
- New
|
|
228
|
-
-
|
|
229
|
-
-
|
|
230
|
-
- Modified styling approach
|
|
230
|
+
- New style manifest files or token definitions
|
|
231
|
+
- Modified design systems or styling approaches
|
|
232
|
+
- Structural changes in asset organization
|
|
231
233
|
|
|
232
234
|
**How to update:**
|
|
233
235
|
|
|
234
|
-
-
|
|
235
|
-
- Update
|
|
236
|
-
- Maintain
|
|
236
|
+
- Document new styling patterns or tokens
|
|
237
|
+
- Update architecture guides if styling methodology changed
|
|
238
|
+
- Maintain existing documentation for core design principles
|
|
237
239
|
|
|
238
240
|
### Dependencies Detection
|
|
239
241
|
|
|
@@ -0,0 +1,42 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Internal logic for Feature implementation within flow-work
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Feature Logic
|
|
6
|
+
|
|
7
|
+
This file contains the detailed execution logic for implementing new features, imported by `@flow-work.md`.
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
## 🚀 Feature Implementation Flow
|
|
11
|
+
|
|
12
|
+
### 1. Specification (auto-skip if User Story)
|
|
13
|
+
- Extract requirements from User Story or Roadmap.
|
|
14
|
+
- If interactive, ask clarification questions (Multiple Choice).
|
|
15
|
+
- Save to `specs/ai-flow/work/[feature]/spec.md`.
|
|
16
|
+
|
|
17
|
+
### 2. Technical Planning
|
|
18
|
+
- Analyze codebase for patterns.
|
|
19
|
+
- Generate `plan.md` with Fibonacci estimation.
|
|
20
|
+
- Organize tasks into layers (Data, Logic, API, Test, Docs).
|
|
21
|
+
- **User Confirmation Required.**
|
|
22
|
+
|
|
23
|
+
### 3. Progressive Implementation
|
|
24
|
+
Choose mode:
|
|
25
|
+
- **Auto**: Complete all tasks without pausing.
|
|
26
|
+
- **Phase-by-phase**: Pause and validate after each phase.
|
|
27
|
+
- **Task-by-task**: Pause after each task.
|
|
28
|
+
|
|
29
|
+
### 4. Definition of Done (DoD)
|
|
30
|
+
- If HU mode: Validate against Gherkin scenarios.
|
|
31
|
+
- Run tests and linting.
|
|
32
|
+
- Perform security check.
|
|
33
|
+
|
|
34
|
+
---
|
|
35
|
+
## 🌿 Git Branching Strategy
|
|
36
|
+
- Generate slug from name.
|
|
37
|
+
- Execute `git checkout -b feature/[slug]`.
|
|
38
|
+
- Maintain `status.json` with commit history.
|
|
39
|
+
|
|
40
|
+
---
|
|
41
|
+
## 📦 status.json Persistence
|
|
42
|
+
Ensure `progress`, `git`, and `metadata` sections are always updated.
|
|
@@ -0,0 +1,33 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Internal logic for Bug Fixes within flow-work
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Fix Logic
|
|
6
|
+
|
|
7
|
+
This file contains the detailed execution logic for bug fixes, imported by `@flow-work.md`.
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
## 🔧 Bug Fix Workflow
|
|
11
|
+
|
|
12
|
+
### 1. Classification
|
|
13
|
+
- **QUICK**: 1 file, obvious cause (typo, null check).
|
|
14
|
+
- **COMPLEX**: Multiple files, investigation needed.
|
|
15
|
+
|
|
16
|
+
### 2. Deep Analysis (Complex Fix)
|
|
17
|
+
- Create work directory.
|
|
18
|
+
- Document root cause in `analysis.md`.
|
|
19
|
+
- Map side effects.
|
|
20
|
+
|
|
21
|
+
### 3. Implementation
|
|
22
|
+
- **Test-First Approach**: Add a failing test case that reproduces the bug before fixing it.
|
|
23
|
+
- Apply the fix.
|
|
24
|
+
- Verify tests pass.
|
|
25
|
+
|
|
26
|
+
---
|
|
27
|
+
## 🌿 Git Branching Strategy
|
|
28
|
+
- **QUICK**: Usually work on current branch.
|
|
29
|
+
- **COMPLEX**: Execute `git checkout -b fix/[slug]`.
|
|
30
|
+
|
|
31
|
+
---
|
|
32
|
+
## 📦 status.json Persistence
|
|
33
|
+
Track root cause details and verification test status.
|
|
@@ -0,0 +1,32 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Internal logic for Refactor implementation within flow-work
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Refactor Logic
|
|
6
|
+
|
|
7
|
+
This file contains the detailed execution logic for code refactoring, imported by `@flow-work.md`.
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
## 🔄 Refactoring Workflow
|
|
11
|
+
|
|
12
|
+
### 1. Scope Identification
|
|
13
|
+
- Map affected files and dependencies.
|
|
14
|
+
- Confirm no behavior change is expected.
|
|
15
|
+
- Validate against architecture patterns.
|
|
16
|
+
|
|
17
|
+
### 2. Implementation Strategy
|
|
18
|
+
- Use `plan.md` to map extraction/renaming/moving steps.
|
|
19
|
+
- Set "No behavior change" as the primary constraint.
|
|
20
|
+
|
|
21
|
+
### 3. Execution & Safety
|
|
22
|
+
- Update imports and references across the codebase.
|
|
23
|
+
- **Critical**: Existing tests must pass without modification (unless test itself is refactored).
|
|
24
|
+
- Run `/flow-check` to verify no regressions.
|
|
25
|
+
|
|
26
|
+
---
|
|
27
|
+
## 🌿 Git Branching Strategy
|
|
28
|
+
- Execute `git checkout -b refactor/[slug]`.
|
|
29
|
+
|
|
30
|
+
---
|
|
31
|
+
## 📦 status.json Persistence
|
|
32
|
+
Track `percentage` of updated occurrences and updated file paths.
|
|
@@ -0,0 +1,32 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Internal logic for Resuming work within flow-work
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Resume Logic
|
|
6
|
+
|
|
7
|
+
This file contains the logic for detecting and resuming paused work items, imported by `@flow-work.md`.
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
## ⏸️ Resume Workflow
|
|
11
|
+
|
|
12
|
+
### 1. Work Detection
|
|
13
|
+
- Scan `specs/ai-flow/work/` for any directories.
|
|
14
|
+
- Group by type and last updated timestamp.
|
|
15
|
+
|
|
16
|
+
### 2. Selection Menu
|
|
17
|
+
- Show active tasks with percentage and current branch.
|
|
18
|
+
- Identify current Git branch and match with work items (⭐⭐ marker).
|
|
19
|
+
|
|
20
|
+
### 3. Context Restoration
|
|
21
|
+
- Load `spec.md`, `plan.md`, `task.md`, and `status.json`.
|
|
22
|
+
- Verify Git branch:
|
|
23
|
+
- If mismatch: Prompt to switch to `git.branchName`.
|
|
24
|
+
- Identify the first incomplete task in `task.md`.
|
|
25
|
+
|
|
26
|
+
### 4. Implementation Continuation
|
|
27
|
+
- Resume with the previously saved `implementationMode`.
|
|
28
|
+
- Continue execution from where it was paused.
|
|
29
|
+
|
|
30
|
+
---
|
|
31
|
+
## 📦 status.json Persistence
|
|
32
|
+
Update `timestamps.lastUpdated` upon resuming.
|
|
@@ -0,0 +1,127 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Central Orchestrator for Feature, Refactor, and Fix workflows
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# AI Flow - Unified Work Orchestrator
|
|
6
|
+
|
|
7
|
+
**YOU ARE AN EXPERT SOFTWARE ARCHITECT AND WORKFLOW COORDINATOR.**
|
|
8
|
+
|
|
9
|
+
Your mission is to orchestrate development tasks through an interactive workflow when the user executes `/flow-work`.
|
|
10
|
+
|
|
11
|
+
---
|
|
12
|
+
## Command: `/flow-work`
|
|
13
|
+
|
|
14
|
+
### Objective
|
|
15
|
+
Provide a single, intelligent entry point for all development work (New Features, Refactorings, and Bug Fixes) with automatic context detection and interactive planning.
|
|
16
|
+
|
|
17
|
+
### Usage Modes
|
|
18
|
+
- **`/flow-work`** → Resume paused work (if exists) or Interactive mode.
|
|
19
|
+
- **`/flow-work [description]`** → Semantic detection (Feature, Refactor, or Fix).
|
|
20
|
+
- **`/flow-work HU-XXX-XXX`** → Implement specific User Story.
|
|
21
|
+
- **`/flow-work [Feature Name]`** → Implement feature from roadmap.md.
|
|
22
|
+
|
|
23
|
+
---
|
|
24
|
+
## Phase 0: Detection & Strategy (Automatic)
|
|
25
|
+
|
|
26
|
+
**1. Semantic Analysis of Input:**
|
|
27
|
+
|
|
28
|
+
| Input Pattern | Mode | Source / Action |
|
|
29
|
+
|---------------|------|-----------------|
|
|
30
|
+
| `HU-\d{3}-\d{3}` | `USER_STORY` | Load from `docs/user-stories/**/HU-XXX-XXX.md` |
|
|
31
|
+
| `EP-\d{3}` | `EPIC` | Analyze/List User Stories for Epic `EP-XXX` |
|
|
32
|
+
| `T\d{3}(-T\d{3})?` | `TASKS` | Target specific task or range (e.g., `T025-T030`) |
|
|
33
|
+
| `HU-XXX-XXX TXXX-TXXX`| `STORY_TASKS` | Targeted tasks within a specific User Story |
|
|
34
|
+
| Matches `docs/roadmap.md` | `ROADMAP_FEATURE`| Extract section from `docs/roadmap.md` (Partial matches allowed) |
|
|
35
|
+
| "refactor", "move", "extract" | `REFACTOR` | Use `flow-work-refactor.md` |
|
|
36
|
+
| "fix", "bug", "error", "falla" | `FIX` | Detect complexity (Quick vs Complex) |
|
|
37
|
+
| "implement", "create", "new" | `FEATURE` | Use `flow-work-feature.md` |
|
|
38
|
+
| No arguments | `RESUME` | Search for paused work in `specs/ai-flow/work/` |
|
|
39
|
+
|
|
40
|
+
**2. Detection Logic Details:**
|
|
41
|
+
- **USER_STORY / EPIC**: Load metadata from `docs/user-stories/` or `docs/roadmap.md`.
|
|
42
|
+
- **ROADMAP_FEATURE**: Fuzzy search in `docs/roadmap.md` for titles like "User Management" or "Feature 2.2".
|
|
43
|
+
- **TASK RANGES**: If `T025-T030` is provided, find the parent Story or Feature in current context or roadmap.
|
|
44
|
+
- **SIMPLE FIX**: Affects 1 file, obvious cause, <10 lines fix. → Use `flow-work-fix.md` (Quick).
|
|
45
|
+
- **COMPLEX FIX**: Multi-file, architectural, performance/security. → Use `flow-work-fix.md` (Deep).
|
|
46
|
+
|
|
47
|
+
---
|
|
48
|
+
## Phase 1: Interactive Analysis
|
|
49
|
+
|
|
50
|
+
**1. Context Loading (Multi-Source):**
|
|
51
|
+
**CRITICAL**: Regardless of whether a `USER_STORY` ID or a `ROADMAP_FEATURE` name is provided, you MUST attempt to load context from **BOTH** sources:
|
|
52
|
+
- **`docs/roadmap.md`**: To understand high-level scope, epic relationships, and technical dependencies.
|
|
53
|
+
- **`docs/user-stories/**/HU-XXX-XXX.md`**: To get granular details (Acceptance Criteria, Gherkin Scenarios, QA cases).
|
|
54
|
+
|
|
55
|
+
**2. Interactive Questions:**
|
|
56
|
+
- IF both sources provide 100% clarity: Skip questions.
|
|
57
|
+
- IF there is missing info or ambiguity: Ask 3-5 key questions with **Multiple Choice Options** and **Defaults (marked with ⭐)**.
|
|
58
|
+
|
|
59
|
+
**Example Interaction:**
|
|
60
|
+
> 📝 I need to clarify some details for this feature:
|
|
61
|
+
> 1. What authentication provider should we use? [default: A]
|
|
62
|
+
> A) JWT (Local) ⭐
|
|
63
|
+
> B) OAuth2 (Google/GitHub)
|
|
64
|
+
> C) Firebase Auth
|
|
65
|
+
>
|
|
66
|
+
> 2. Should we implement audit logs for this? [default: B]
|
|
67
|
+
> A) Yes
|
|
68
|
+
> B) No ⭐
|
|
69
|
+
|
|
70
|
+
---
|
|
71
|
+
## Phase 2: Planning & Documentation
|
|
72
|
+
|
|
73
|
+
1. **`spec.md`**: Generate/Update in `specs/ai-flow/work/[task-name]/spec.md`.
|
|
74
|
+
- Ask for user approval.
|
|
75
|
+
2. **`plan.md`**: Generate technical approach and task list.
|
|
76
|
+
- Assign Feature Number (NNN).
|
|
77
|
+
- Story Points estimation (Fibonacci).
|
|
78
|
+
- Phase organization.
|
|
79
|
+
- Ask for user approval.
|
|
80
|
+
3. **`task.md`**: Generate the checklist of tactical tasks.
|
|
81
|
+
4. **`status.json`**: Initialize/Update metadata (progress, branch, validation state).
|
|
82
|
+
|
|
83
|
+
---
|
|
84
|
+
## Phase 3: Execution (Branch Creation)
|
|
85
|
+
|
|
86
|
+
**Upon confirmation to start implementation:**
|
|
87
|
+
|
|
88
|
+
1. **Generate Branch Name**:
|
|
89
|
+
- `feature/[slug]`
|
|
90
|
+
- `refactor/[slug]`
|
|
91
|
+
- `fix/[slug]`
|
|
92
|
+
2. **Execute**: `git checkout -b [branch-name]`.
|
|
93
|
+
3. **Update `status.json`**: Record branch name and start timestamp.
|
|
94
|
+
4. **Implementation**: Proceed according to the selected mode (Auto, Phase-by-phase, Task-by-task).
|
|
95
|
+
|
|
96
|
+
---
|
|
97
|
+
## Phase 4: Finalization & Archiving
|
|
98
|
+
|
|
99
|
+
**When all tasks in `task.md` are complete (✅) and validated:**
|
|
100
|
+
|
|
101
|
+
1. **Sugerir Próximos Pasos**:
|
|
102
|
+
- **`/flow-check`**: Ejecutar tests y revisión de código combinada.
|
|
103
|
+
- **`/flow-docs-sync`**: Sincronizar la documentación técnica.
|
|
104
|
+
- **`/flow-commit`**: Crear commits atómicos.
|
|
105
|
+
|
|
106
|
+
2. **Proceso de Archivado (Automático tras aprobación)**:
|
|
107
|
+
- Una vez el usuario confirma que el trabajo está listo para ser cerrado:
|
|
108
|
+
- **Mover**: `specs/ai-flow/work/[task-name]/` → `specs/ai-flow/archive/YYYY-MM/[task-name]/`.
|
|
109
|
+
- **Actualizar `status.json`**: Cambiar `status` a `"COMPLETED"` y registrar `timestamps.completed`.
|
|
110
|
+
- **Cleanup**: Mantener limpia la carpeta `work` para que `/flow-work` detecte solo tareas activas.
|
|
111
|
+
|
|
112
|
+
3. **Resumen Final**:
|
|
113
|
+
- Mostrar estadísticas finales de tiempo, archivos y cobertura antes de archivar.
|
|
114
|
+
|
|
115
|
+
---
|
|
116
|
+
## Orchestration Rules
|
|
117
|
+
|
|
118
|
+
- **DRY Logic**: This file handles the high-level orchestration.
|
|
119
|
+
- **Delegation**:
|
|
120
|
+
- Detailed Feature logic → `@flow-work-feature.md`
|
|
121
|
+
- Detailed Refactor logic → `@flow-work-refactor.md`
|
|
122
|
+
- Detailed Fix logic → `@flow-work-fix.md`
|
|
123
|
+
- Resume logic → `@flow-work-resume.md`
|
|
124
|
+
- **State Persistence**: Always read/write to `specs/ai-flow/work/[name]/status.json` to maintain state across sessions.
|
|
125
|
+
|
|
126
|
+
---
|
|
127
|
+
**BEGIN EXECUTION when user runs `/flow-work [args]`**
|