ai-flow-dev 2.1.8 → 2.2.0

package/prompts/backend/flow-build-phase-1.md

@@ -2,6 +2,13 @@
 
 > **Order for this phase:** 1.1 → 1.2 → 1.3 → 1.4 → 1.5 → 1.6 → 1.7 → 1.8 → 1.9 → 1.10
 
+> **📌 Scope-based behavior:**
+> - **MVP/Basic Scope:** Focus only on core requirements. Skip advanced business logic questions.
+> - **Production-Ready Scope:** In-depth exploration of compliance, scalability, and long-term business goals.
+
+### Objective
+Define the project's core purpose, business rules, and high-level requirements to ensure the foundation is solid before technical design begins.
+
 > **📌 Note:** If Phase 0 was executed, some questions may already be answered. Skip those and only ask what's missing.
 
 **1.1 Project Name & Description (with Smart Refinement)**
@@ -573,10 +580,25 @@ The document has been created with all the information from Phase 1.
 **If user edits the file:**
 Execute `read_file('project-brief.md')` to refresh context before continuing.
 ---
-**Proceed to Phase 2 after document is generated and optionally reviewed.**
 
-> ⚠️ **CRITICAL:** DO NOT generate README.md in Phase 1. README.md is ONLY generated in Phase 8 (step 8.5) after framework initialization.
+## 📝 Generated Documents
+
+After Phase 1, generate/update:
+- `project-brief.md` - Core project discovery and requirements
+
+---
+
+**Next Phase:** Phase 2 - Data Architecture (15-20 min)
+
+Read: `.ai-flow/prompts/backend/flow-build-phase-2.md`
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+
 ---
+
 ## PHASE 2: Data Architecture (15-20 min)
 
 ````

package/prompts/backend/flow-build-phase-10.md

@@ -1,15 +1,12 @@
-# AI Flow - Phase 10: User Stories Generation
+## PHASE 10: Agile Planning - User Stories (5-10 min)
 
-**YOU ARE AN EXPERT PRODUCT OWNER AND AGILE SPECIALIST.**
+> **Order for this phase:** OPTIONAL. Executed after Phase 9 or on demand.
 
-Generate detailed User Stories from the roadmap created in Phase 9.
+> **📌 Scope-based behavior:**
+> - **All Scopes:** Generate detailed User Stories for each milestone in the Roadmap.
 
-## Prerequisites
-
-- Phase 9 completed (`docs/roadmap.md` exists with context variables)
-- All Phase 1-8 documentation exists
-
-## Duration
+### Objective
+Break down technical roadmap tasks into user-centric Agile requirements (User Stories) to facilitate development, testing, and acceptance criteria definition.
 
 - All Epics: 30-60 minutes
 - One Epic: 5-10 minutes
@@ -29,6 +26,7 @@ Generate detailed User Stories from the roadmap created in Phase 9.
 
 ## Workflow: 4 Steps
 
+// turbo
 ### Step 10.1: Load Context from Roadmap
 
 **Extract from `docs/roadmap.md`:**
@@ -321,3 +319,22 @@ Next steps:
 2. Start implementing: /flow-dev-feature HU-001-001
 3. Generate more: /flow-build fase 10 EP-XXX
 ```
+
+---
+
+## 📝 Generated Documents
+
+After Phase 10, generate/update:
+
+- `docs/user-stories/*.md` - Detailed Agile requirements
+- `docs/roadmap.md` - (updated with story links)
+
+---
+
+**CONGRATULATIONS!** Project documentation and planning are complete. 🚀
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+

package/prompts/backend/flow-build-phase-2.md

@@ -2,7 +2,9 @@
 
 > **Order for this phase:** 2.1 → 2.2 → 2.3 → 2.4 → 2.5 → 2.6 → 2.7
 
-> **📌 Note:** If Phase 0 detected database/entities from code, those will be pre-filled. Review and confirm.
+> **📌 Scope-based behavior:**
+> - **MVP/Basic Scope:** Focus only on essential entities and basic relationships.
+> - **Production-Ready Scope:** In-depth modeling including indexes, constraints, and audit fields.
 
 ### Objective
 
@@ -531,6 +533,22 @@ Execute `read_file('docs/data-model.md')` to refresh context before continuing.
 
 ---
 
-## **Proceed to Phase 3 after document is generated and optionally reviewed.**
+## 📝 Generated Documents
+
+After Phase 2, generate/update:
+- `docs/data-model.md` - Database schema and entity relationships
+
+---
+
+**Next Phase:** Phase 3 - System Architecture (15-20 min)
+
+Read: `.ai-flow/prompts/backend/flow-build-phase-3.md`
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+
+---
 
 ## PHASE 3: System Architecture (15-20 min)

package/prompts/backend/flow-build-phase-3.md

@@ -741,4 +741,23 @@ Execute `read_file()` for both documents to refresh context before continuing.
 
 ---
 
+## 📝 Generated Documents
+
+After Phase 3, generate/update:
+- `docs/architecture.md` - Technical stack and patterns
+- `ai-instructions.md` - Instructions for AI agents
+
+---
+
+**Next Phase:** Phase 4 - Security & Authentication (15-20 min)
+
+Read: `.ai-flow/prompts/backend/flow-build-phase-4.md`
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+
+---
+
 ## PHASE 4: Security & Authentication (15-20 min)

package/prompts/backend/flow-build-phase-4.md

@@ -461,7 +461,7 @@ Input Validation: [strategy + sanitization rules + size limits + file upload val
 
 Is this correct? (Yes/No)
 ```
----
+---
 ### 📄 Generate Phase 4 Documents
 
 **Before starting generation:**
@@ -501,12 +501,35 @@ Documents have been created with all Phase 4 information.
 
 **If user edits files:**
 Re-read files to refresh context before continuing.
----
+---
 **Proceed to Phase 5 only after documents are validated.**
 
 > ⚠️ **CRITICAL:** DO NOT generate README.md in this phase. README.md is ONLY generated in Phase 8 (step 8.5) after framework initialization.
----
+---
+---
+
+---
+
+## 📝 Generated Documents
+
+After Phase 4, generate/update:
+- `specs/security.md` - Security policies and authentication details
+
+---
+
+**Next Phase:** Phase 5 - Development Standards (15-20 min)
+
+Read: `.ai-flow/prompts/backend/flow-build-phase-5.md`
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+
+---
+
 ## PHASE 5: Development Standards (15-20 min)
 
 
 
+

package/prompts/backend/flow-build-phase-5.md

@@ -662,4 +662,23 @@ Re-read files to refresh context before continuing.
 
 ---
 
-## PHASE 6: Testing Strategy (10 min)
+## 📝 Generated Documents
+
+After Phase 5, generate/update:
+- `docs/code-standards.md` - Development practices and quality rules
+
+---
+
+**Next Phase:** Phase 6 - Testing Strategy (15-20 min)
+
+Read: `.ai-flow/prompts/backend/flow-build-phase-6.md`
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+
+---
+
+## PHASE 6: Testing Strategy (15-20 min)
+

package/prompts/backend/flow-build-phase-6.md

@@ -1,4 +1,4 @@
-## PHASE 6: Testing Strategy (15-25 min)
+## PHASE 6: Testing Strategy (15-20 min)
 
 > **Order for this phase:**
 >
@@ -481,5 +481,25 @@ Re-read file to refresh context before continuing.
 
 > ⚠️ **CRITICAL:** DO NOT generate README.md in this phase. README.md is ONLY generated in Phase 8 (step 8.5) after framework initialization.
 ---
-## PHASE 7: Operations & Deployment (10 min)
+
+## 📝 Generated Documents
+
+After Phase 6, generate/update:
+- `docs/testing.md` - Testing strategy and quality gates
+
+---
+
+**Next Phase:** Phase 7 - Operations & Deployment (10-15 min)
+
+Read: `.ai-flow/prompts/backend/flow-build-phase-7.md`
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+
+---
+
+## PHASE 7: Operations & Deployment (10-15 min)
+
 ````

package/prompts/backend/flow-build-phase-7.md

@@ -1,12 +1,12 @@
-## PHASE 7: Operations & Deployment (10 min)
+## PHASE 7: Operations & Deployment (10-15 min)
 
-> **Order for this phase:** 7.1 → 7.2 → 7.3 → 7.4 → 7.4.1 → 7.5 → 7.6 → 7.7 → 7.7.1 → 7.7.2 → 7.8 → 7.9 → 7.9.1 → 7.9.2 → 7.9.3 → 7.9.4 → 7.10 → 7.11
+> **Order for this phase:** 7.1 → 7.2 → 7.3 → 7.4 → 7.4.1 → 7.5 → 7.6 → 7.7 → 7.7.1 → 7.7.2 → 7.8 → 7.9 → 7.9.1 → 7.9.2 → 7.9.3 → 7.9.4 → 7.10
 
 > **📌 Scope-based behavior:**
 >
-> - **MVP:** Ask 7.1-7.4 only (deployment basics), skip 7.5-7.11 (monitoring, scaling, backups), mark as "TBD"
-> - **Production-Ready:** Ask 7.1-7.8, simplify 7.9-7.11 (advanced monitoring and scaling)
-> - **Enterprise:** Ask all questions 7.1-7.11 with emphasis on reliability and disaster recovery
+> - **MVP:** Ask 7.1-7.4 only (deployment basics), skip 7.5-7.10 (monitoring, scaling, backups), mark as "TBD"
+> - **Production-Ready:** Ask 7.1-7.8, simplify 7.9-7.10 (advanced monitoring and resilience)
+> - **Enterprise:** Ask all questions 7.1-7.10 with emphasis on reliability and disaster recovery
 
 ### Objective
 
@@ -943,279 +943,40 @@ Re-read files to refresh context before continuing.
 
 ---
 
-## **Proceed to final generation only after documents are validated.**
-
-## FINAL STEP: Generate Remaining Documents & Tool Configs
-
-### ⏸️ CHECKPOINT: Final Document Review
-
-```
-⏸️ CHECKPOINT: Review Generated Documents
-
-We have successfully generated the following documents across all phases:
-
-✅ project-brief.md (Phase 1)
-✅ docs/data-model.md (Phase 2)
-✅ docs/architecture.md (Phase 3)
-✅ ai-instructions.md (Phase 3)
-✅ specs/security.md (Phase 4)
-✅ docs/code-standards.md (Phase 5)
-✅ docs/testing.md (Phase 6)
-✅ docs/operations.md (Phase 7)
-✅ specs/configuration.md (Phase 7)
-✅ .env.example (Phase 7)
-
-Before generating the final documents (AGENT.md, README.md, API docs, etc.),
-do you want to review or edit any of these documents?
-
-A) ✅ All documents are correct, proceed with final generation
-B) 📝 I need to review/edit some documents (I'll wait)
-```
-
-**If user selects B:**
-
-```
-Perfect. Please review and edit any documents you need to change.
-When you're done, type "ready" and I'll re-read ALL documents
-to ensure the final generation uses the most up-to-date information.
-```
-
-## Then execute `read_file()` for ALL previously generated documents.
-
-### 📖 Loading Complete Context for Final Generation
-
-```
-📖 Re-reading all generated documents to ensure accuracy...
-
-✅ Re-reading project-brief.md
-✅ Re-reading docs/data-model.md
-✅ Re-reading docs/architecture.md
-✅ Re-reading ai-instructions.md
-✅ Re-reading specs/security.md
-✅ Re-reading docs/code-standards.md
-✅ Re-reading docs/testing.md
-✅ Re-reading docs/operations.md
-✅ Re-reading specs/configuration.md
-✅ Re-reading .env.example
-
-✅ Context fully loaded and updated!
-
-🎉 Now generating final 5 documents:
-
-1. docs/business-flows.md - Business process flows and mermaid diagrams
-2. docs/api.md - API endpoints and conventions reference
-3. docs/contributing.md - Contribution guidelines
-4. README.md - Project overview (requires all previous context)
-5. AGENT.md - Universal AI configuration (aggregates all docs)
-
-Plus tool-specific configurations:
-- .clauderules (if Claude selected)
-- .cursorrules (if Cursor selected)
-- .github/copilot-instructions.md (if Copilot selected)
-
-Generating...
-```
-
-### Transition to Phase 8
-
-All information from Phases 1-7 has been collected and documented.
-
-Phase 8 will handle:
-
-- Project state detection
-- Framework initialization (if needed)
-- Final documentation generation
-- README.md creation with intelligent merge
-- Tool-specific configurations
-
-**Phase 8 will be executed next to complete the AI Flow process.**
+### Phase 7 Completion
 
 ```
 ✅ Phase 7 Complete!
 
-Generated 3 documents successfully:
-
-Phase 7:
+Generated documents:
 ✅ docs/operations.md
 ✅ specs/configuration.md
 ✅ .env.example
----
-📊 Progress Summary:
-
-Phases 1-7 completed:
-✅ project-brief.md (Phase 1)
-✅ docs/data-model.md (Phase 2)
-✅ docs/architecture.md (Phase 3)
-✅ ai-instructions.md (Phase 3)
-✅ specs/security.md (Phase 4)
-✅ docs/code-standards.md (Phase 5)
-✅ docs/testing.md (Phase 6)
-✅ docs/operations.md (Phase 7)
-✅ specs/configuration.md (Phase 7)
-✅ .env.example (Phase 7)
-
-Remaining:
-⏭️  Phase 8: Project setup & final documentation
----
-🎯 Next: Phase 8 - Project Setup & Final Documentation
-
-Phase 8 will:
-- 🔍 Detect project state (new vs existing)
-- 🚀 Initialize framework (optional, for new projects)
-- 📝 Generate final docs (business-flows.md, api.md, contributing.md)
-- 📖 Generate AGENT.md (master documentation index)
-- 📄 Generate README.md (with intelligent merge if needed)
-- 🤖 Create tool-specific configs (based on AI selection)
 
-> ⚠️ **CRITICAL:** README.md MUST NOT exist yet. It will be generated in Phase 8 step 8.5 AFTER framework initialization to avoid conflicts.
+📝 Would you like to review these documents before proceeding to Phase 8?
 
-Continue to Phase 8? (yes/no)
+→ If yes: Edit the files and type "ready" when done.
+→ If no: Type "continue" to proceed to Phase 8.
 ```
 
 ---
 
-## EXECUTION CHECKLIST FOR AI ASSISTANT
-
-When executing this master prompt:
-
-**PHASE 0 (Existing Projects Only):**
-
-- [ ] Check if project has existing code/documentation
-- [ ] Search for AI instruction files (copilot-instructions.md, .clauderules, .cursorrules, AGENT.md, etc.)
-- [ ] Search for README.md, package.json, and config files
-- [ ] **Perform deep code analysis:**
-  - [ ] Recursively scan source directories (src/, app/, lib/, modules/)
-  - [ ] Count files by type (controllers, services, entities, DTOs, etc.)
-  - [ ] Identify architecture pattern (feature-based, layer-based, modular, hybrid)
-  - [ ] **Parse source code files for AST-based analysis:**
-    - [ ] Detect decorators (@Controller, @Get, @Post for NestJS; @app.get() for FastAPI)
-    - [ ] Extract API endpoints (method, path, parameters, middleware)
-    - [ ] Identify services and dependency injection patterns
-    - [ ] Detect error handling patterns
-  - [ ] **Analyze database schemas:**
-    - [ ] Parse Prisma schema.prisma (models, relationships, indexes)
-    - [ ] Parse TypeORM entities (@Entity, @Column, relationships)
-    - [ ] Parse SQLAlchemy models
-    - [ ] Extract complete entity definitions with fields and relationships
-  - [ ] **Analyze dependencies:**
-    - [ ] Extract exact versions from package.json/requirements.txt/etc.
-    - [ ] Identify framework, ORM, validation, auth, testing libraries
-    - [ ] Detect security-related packages
-  - [ ] **Detect code patterns:**
-    - [ ] Dependency injection usage
-    - [ ] Error handling implementation
-    - [ ] Validation patterns (DTOs, schemas)
-    - [ ] Logging setup
-    - [ ] Testing configuration
-    - [ ] Caching strategies
-    - [ ] Background job systems
-- [ ] Present detailed detection results (0.2) including:
-  - [ ] Architecture pattern detected
-  - [ ] Complete list of API endpoints with details
-  - [ ] All entities with schemas and relationships
-  - [ ] Dependencies with versions
-  - [ ] Code quality indicators
-- [ ] **Generate proactive suggestions (0.2.1):**
-  - [ ] Analyze code for security gaps (validation, rate limiting, CORS, etc.)
-  - [ ] Identify architecture improvements
-  - [ ] Detect documentation gaps
-  - [ ] Suggest testing improvements
-  - [ ] Recommend performance optimizations
-  - [ ] Present suggestions with actions and impacts
-- [ ] **Offer export option (0.3):**
-  - [ ] Ask if user wants to export analysis to JSON
-  - [ ] If yes, create .ai-flow/cache/docs-analysis.json with complete analysis
-- [ ] Let user choose: A) Use detected info, B) Start fresh, C) Review/edit detected info
-- [ ] **Handle suggestions:**
-  - [ ] If user selected A (address during build), integrate suggestions into relevant phases
-  - [ ] If user selected B (save for later), create suggestions.md file
-  - [ ] If user selected C (skip), continue without changes
-- [ ] Pre-populate answers based on detected information
-- [ ] Mark questions that still need answers
-
-**PHASES 1-7 (All Projects):**
-
-- [ ] Execute phases 1-7 in exact order
-- [ ] **SKIP questions already answered** from Phase 0 detection (existing projects)
-- [ ] Ask remaining questions ONE BY ONE (wait for answer)
-- [ ] Wait for user response before proceeding
-- [ ] Provide recommendations (⭐🔥⚡🏆)
-- [ ] Offer multiple choice where applicable
-- [ ] Summarize each phase for confirmation
-- [ ] Collect ALL required information
-- [ ] Generate documents **incrementally** after each phase
-- [ ] Ask for user validation after each document generation
-- [ ] Re-read generated documents at start of each new phase
-- [ ] Perform final checkpoint before generating AGENT.md and README.md
-- [ ] Re-read ALL documents before final generation to ensure accuracy
-- [ ] Use templates from `.ai-flow/templates/`
-- [ ] Create tool-specific configs based on AI selection
-- [ ] Validate no placeholders remain
-- [ ] Provide clear next steps
-
-**DO NOT:**
-
-- ❌ Skip Phase 0 detection for existing projects
-- ❌ Skip deep code analysis - always perform comprehensive AST-based parsing when code exists
-- ❌ Present superficial analysis - include detailed endpoints, entities, and patterns
-- ❌ Ignore proactive suggestions - always generate and present improvement opportunities
-- ❌ Ask questions already answered by detected files or code analysis
-- ❌ Ignore existing AI instruction files
-- ❌ Skip questions or phases
-- ❌ Assume answers without asking (when info is not detected)
-- ❌ Generate ALL documents at the end (generate incrementally!)
-- ❌ Skip document validation after generation
-- ❌ Forget to re-read documents before using their info
-- ❌ Generate final documents without re-reading all previous docs
-- ❌ Leave placeholder text in final documents
-- ❌ Skip exporting analysis JSON if user requested it
-- ❌ Rush through the analysis - thoroughness saves time later
-
-**ESTIMATED TIME:**
-
-**New Projects:**
-
-- Phase 1: 15-20 min
-- Phase 2: 15-20 min
-- Phase 3: 15-20 min
-- Phase 4: 15-20 min
-- Phase 5: 15-20 min
-- Phase 6: 10 min
-- Phase 7: 10 min
-- **Total: 90-120 minutes**
-
-**Existing Projects (with Phase 0 detection):**
-
-- Phase 0: 10-20 min (deep code analysis + suggestions + review)
-  - File detection: 2-3 min
-  - Deep code analysis (AST parsing, schema extraction): 5-10 min
-  - Generating suggestions: 2-3 min
-  - User review and export: 1-4 min
-- Phases 1-7: 30-60 min (only missing questions)
-- **Total: 40-80 minutes** (40-60% time saved!)
-
-## This is an investment that will save 10-20 hours over the project lifecycle.
-
-**EXECUTION FLOW:**
-
-1. **START:** User runs `/flow-build`
-2. **DETECT:** Check for existing project files (Phase 0)
-   - If existing files found → Run Phase 0 deep analysis:
-     - 0.1: Deep Code Analysis (file structure, AST parsing, schema extraction, pattern detection)
-     - 0.2: Present Detailed Detection Results (architecture, endpoints, entities, dependencies, code quality)
-     - 0.2.1: Generate Proactive Suggestions (security, architecture, documentation, testing, performance)
-     - 0.3: Export Analysis Results (optional JSON export)
-     - 0.4: Load Existing Context (pre-populate answers)
-   - If no files found → Skip to Phase 1
-3. **EXECUTE:** Run Phases 1-7 with pre-populated answers (if any)
-   - Integrate suggestions from Phase 0 into relevant phases if user selected option A
-4. **GENERATE:** Create documents incrementally with validation
-5. **COMPLETE:** Final checkpoint and remaining documents
+## 📝 Generated Documents
+
+After Phase 7, generate/update:
+
+- `docs/operations.md` - Operations and deployment guide
+- `specs/configuration.md` - Configuration specification  
+- `.env.example` - Environment variables template
 
 ---
 
-**BEGIN EXECUTION when user runs `/flow-build`**
+**Next Phase:** Phase 8 - Project Setup & Final Documentation
 
-```
+Read: `.ai-flow/prompts/backend/flow-build-phase-8.md`
 
-```
+---
+
+**Last Updated:** 2025-12-20
+
+**Version:** 2.1.8