mdan-cli 2.5.1 → 2.7.0
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.
- package/AGENTS.md +76 -1
- package/README.md +274 -4
- package/agents/auto-orchestrator.md +343 -0
- package/agents/devops.md +511 -94
- package/cli/mdan.py +111 -6
- package/cli/mdan_crewai.py +539 -0
- package/core/crewai_orchestrator.md +419 -0
- package/core/debate-protocol.md +454 -0
- package/core/universal-envelope.md +113 -0
- package/integrations/__init__.py +33 -0
- package/integrations/crewai/__init__.py +27 -0
- package/integrations/crewai/agents/__init__.py +21 -0
- package/integrations/crewai/agents/architect_agent.py +264 -0
- package/integrations/crewai/agents/dev_agent.py +271 -0
- package/integrations/crewai/agents/devops_agent.py +421 -0
- package/integrations/crewai/agents/doc_agent.py +388 -0
- package/integrations/crewai/agents/product_agent.py +203 -0
- package/integrations/crewai/agents/security_agent.py +386 -0
- package/integrations/crewai/agents/test_agent.py +358 -0
- package/integrations/crewai/agents/ux_agent.py +257 -0
- package/integrations/crewai/flows/__init__.py +13 -0
- package/integrations/crewai/flows/auto_flow.py +451 -0
- package/integrations/crewai/flows/build_flow.py +297 -0
- package/integrations/crewai/flows/debate_flow.py +422 -0
- package/integrations/crewai/flows/discovery_flow.py +267 -0
- package/integrations/crewai/orchestrator.py +558 -0
- package/integrations/crewai/skills/__init__.py +8 -0
- package/integrations/crewai/skills/skill_router.py +534 -0
- package/integrations/crewai/tools/__init__.py +11 -0
- package/integrations/crewai/tools/file_tool.py +355 -0
- package/integrations/crewai/tools/serper_tool.py +169 -0
- package/integrations/crewai/tools/sql_tool.py +435 -0
- package/memory/CONTEXT-SAVE-FORMAT.md +328 -0
- package/memory/MEMORY-AUTO.json +66 -0
- package/memory/RESUME-PROTOCOL.md +379 -0
- package/package.json +1 -1
- package/phases/auto-01-load.md +165 -0
- package/phases/auto-02-discover.md +207 -0
- package/phases/auto-03-plan.md +509 -0
- package/phases/auto-04-architect.md +567 -0
- package/phases/auto-05-implement.md +713 -0
- package/phases/auto-06-test.md +559 -0
- package/phases/auto-07-deploy.md +510 -0
- package/phases/auto-08-doc.md +970 -0
- package/skills/azure-devops/skill.md +1757 -0
- package/templates/dotnet-blazor/README.md +415 -0
- package/templates/external-services/ExampleService.cs +361 -0
- package/templates/external-services/IService.cs +113 -0
- package/templates/external-services/README.md +325 -0
- package/templates/external-services/ServiceBase.cs +492 -0
- package/templates/external-services/ServiceProvider.cs +243 -0
- package/templates/prompts/devops-agent.yaml +327 -0
- package/templates/prompts.json +15 -1
- package/templates/sql-server/README.md +37 -0
- package/templates/sql-server/functions.sql +158 -0
- package/templates/sql-server/schema.sql +188 -0
- package/templates/sql-server/stored-procedures.sql +284 -0
package/AGENTS.md
CHANGED
|
@@ -228,6 +228,8 @@ my-project/
|
|
|
228
228
|
| `mdan init` | Initialize project |
|
|
229
229
|
| `mdan attach` | Add MDAN to existing project |
|
|
230
230
|
| `mdan phase` | Show/run phase |
|
|
231
|
+
| `mdan auto` | Run autonomous full-cycle development |
|
|
232
|
+
| `mdan resume` | Resume from saved context |
|
|
231
233
|
| `mdan mcp init` | Generate MCP config |
|
|
232
234
|
| `mdan prompt list` | List prompts |
|
|
233
235
|
| `mdan test` | Run tests |
|
|
@@ -235,12 +237,85 @@ my-project/
|
|
|
235
237
|
|
|
236
238
|
---
|
|
237
239
|
|
|
240
|
+
## Agent Registry
|
|
241
|
+
|
|
242
|
+
### Core Agents
|
|
243
|
+
|
|
244
|
+
| Agent | Phase | Expertise |
|
|
245
|
+
|-------|-------|-----------|
|
|
246
|
+
| Product Agent (Khalil) | DISCOVER | Requirements gathering, PRD creation |
|
|
247
|
+
| Architect Agent (Reda) | DESIGN | System architecture, tech stack |
|
|
248
|
+
| UX Agent (Jihane) | DESIGN | User experience, UI/UX design |
|
|
249
|
+
| Dev Agent (Haytame) | BUILD | Implementation, coding |
|
|
250
|
+
| Security Agent (Said) | BUILD/VERIFY | Security review, vulnerability assessment |
|
|
251
|
+
| Test Agent (Youssef) | VERIFY | Testing strategy, test execution |
|
|
252
|
+
| DevOps Agent (Anas) | SHIP | Deployment, CI/CD, Azure |
|
|
253
|
+
| Doc Agent (Amina) | SHIP | Documentation, user guides |
|
|
254
|
+
| Learn Agent | Transverse | Knowledge management, learning |
|
|
255
|
+
|
|
256
|
+
### Autonomous Agents
|
|
257
|
+
|
|
258
|
+
| Agent | Phase | Expertise |
|
|
259
|
+
|-------|-------|-----------|
|
|
260
|
+
| Auto Orchestrator | ALL | Autonomous full-cycle development coordination |
|
|
261
|
+
|
|
262
|
+
The Auto Orchestrator executes all 8 phases (LOAD → DISCOVER → PLAN → ARCHITECT → IMPLEMENT → TEST → DEPLOY → DOC) without human intervention, with context save/load capabilities and multi-agent debate support.
|
|
263
|
+
|
|
264
|
+
---
|
|
265
|
+
|
|
266
|
+
## CrewAI Integration
|
|
267
|
+
|
|
268
|
+
MDANV2 includes a comprehensive CrewAI integration for intelligent multi-agent orchestration.
|
|
269
|
+
|
|
270
|
+
### Installation
|
|
271
|
+
|
|
272
|
+
```bash
|
|
273
|
+
pip install -r requirements_crewai.txt
|
|
274
|
+
```
|
|
275
|
+
|
|
276
|
+
### Environment Variables
|
|
277
|
+
|
|
278
|
+
```bash
|
|
279
|
+
export SERPER_API_KEY="your-serper-api-key"
|
|
280
|
+
export OPENAI_API_KEY="your-openai-api-key"
|
|
281
|
+
```
|
|
282
|
+
|
|
283
|
+
### Usage
|
|
284
|
+
|
|
285
|
+
```python
|
|
286
|
+
import asyncio
|
|
287
|
+
from integrations.crewai import CrewAIOrchestrator
|
|
288
|
+
|
|
289
|
+
async def main():
|
|
290
|
+
orchestrator = CrewAIOrchestrator(
|
|
291
|
+
project_path="/path/to/project",
|
|
292
|
+
llm=your_llm_instance,
|
|
293
|
+
auto_mode=True
|
|
294
|
+
)
|
|
295
|
+
|
|
296
|
+
# Run autonomous mode
|
|
297
|
+
result = await orchestrator.run_auto_mode("Build a todo app")
|
|
298
|
+
|
|
299
|
+
asyncio.run(main())
|
|
300
|
+
```
|
|
301
|
+
|
|
302
|
+
### Components
|
|
303
|
+
|
|
304
|
+
- **8 CrewAI Agents**: Product, Architect, UX, Dev, Test, Security, DevOps, Doc
|
|
305
|
+
- **4 CrewAI Flows**: Auto, Discovery, Build, Debate
|
|
306
|
+
- **3 Custom Tools**: Serper (web search), SQL (async database), File (file operations)
|
|
307
|
+
- **Skill Router**: Intelligent skill detection and routing (50+ skills)
|
|
308
|
+
|
|
309
|
+
### Documentation
|
|
310
|
+
|
|
311
|
+
See `core/crewai_orchestrator.md` for complete documentation.
|
|
312
|
+
|
|
238
313
|
## Resources
|
|
239
314
|
|
|
240
315
|
- [MDAN Documentation](https://github.com/khalilbenaz/MDAN)
|
|
241
316
|
- [Better Agents](https://langwatch.ai/docs/better-agents)
|
|
242
317
|
- [Agent Skills Standard](https://agentskills.io)
|
|
318
|
+
- [CrewAI Documentation](https://docs.crewai.com)
|
|
243
319
|
|
|
244
320
|
---
|
|
245
|
-
|
|
246
321
|
*This file is auto-generated by MDAN. Edit with care.*
|
package/README.md
CHANGED
|
@@ -27,16 +27,21 @@ Les outils IA traditionnels font le travail à votre place, produisant des résu
|
|
|
27
27
|
|
|
28
28
|
---
|
|
29
29
|
|
|
30
|
-
## 🚀 Nouveautés dans MDAN v2.
|
|
30
|
+
## 🚀 Nouveautés dans MDAN v2.5
|
|
31
31
|
|
|
32
32
|
**MDAN évolue rapidement avec des optimisations incluant :**
|
|
33
33
|
|
|
34
|
+
- **MDAN-AUTO v1.0** — Mode de développement autonome sans intervention humaine
|
|
35
|
+
- **Context Save/Resume** — Sauvegarde et reprise automatique du contexte à 80% des tokens
|
|
36
|
+
- **Multi-Agent Debates** — Débats entre agents pour les décisions complexes
|
|
34
37
|
- **Cross Platform Agent Team** — Agents multi-plateformes (Cursor, Windsurf, Claude Code, VS Code)
|
|
35
38
|
- **Sub Agent Inclusion** — Sous-agents spécialisés pour des tâches ciblées
|
|
36
39
|
- **Skills Architecture** — Architecture de skills extensible et modulable
|
|
37
40
|
- **MDAN Builder v1** — Créez vos propres agents et workflows
|
|
38
41
|
- **Dev Loop Automation** — Automatisation complète du cycle de développement
|
|
39
42
|
- **Better Agents Integration** — Scenarios et evaluations intégrés
|
|
43
|
+
- **Templates Génériques** — Templates services externes, .NET Blazor, SQL Server
|
|
44
|
+
- **Architecture Modulaire** — Skills extensibles et modules spécialisés
|
|
40
45
|
|
|
41
46
|
---
|
|
42
47
|
|
|
@@ -119,8 +124,12 @@ MDAN s'étend avec des modules officiels pour domaines spécialisés :
|
|
|
119
124
|
| Module | Description |
|
|
120
125
|
|--------|-------------|
|
|
121
126
|
| **MDAN Core** | Framework principal avec 5 phases |
|
|
127
|
+
| **MDAN-AUTO** | Mode de développement autonome (8 phases) |
|
|
128
|
+
| **External Services** | Templates d'intégration services externes génériques |
|
|
129
|
+
| **.NET Blazor** | Templates projets .NET 8.0 Blazor Server |
|
|
130
|
+
| **SQL Server** | Templates SQL Server (schema, stored procedures, functions) |
|
|
131
|
+
| **Azure DevOps** | Skills pour Azure DevOps Services |
|
|
122
132
|
| **Agile Scrum** | Workflows Agile/Scrum (Sprint, backlog, retrospectives) |
|
|
123
|
-
| **Skills** | Compétences additionnelles |
|
|
124
133
|
|
|
125
134
|
```bash
|
|
126
135
|
# Ajouter un module
|
|
@@ -129,8 +138,83 @@ mdan module add agile-scrum
|
|
|
129
138
|
|
|
130
139
|
---
|
|
131
140
|
|
|
141
|
+
## 📋 Templates Disponibles
|
|
142
|
+
|
|
143
|
+
MDAN fournit des templates génériques et réutilisables pour accélérer le développement :
|
|
144
|
+
|
|
145
|
+
### External Services
|
|
146
|
+
|
|
147
|
+
Templates d'intégration de services externes (APIs, microservices) :
|
|
148
|
+
|
|
149
|
+
- `IService.cs` — Interface de base pour les services
|
|
150
|
+
- `ServiceBase.cs` — Classe de base avec logging et gestion d'erreurs
|
|
151
|
+
- `ServiceProvider.cs` — Gestionnaire de dépendances
|
|
152
|
+
- `ExampleService.cs` — Exemple d'implémentation
|
|
153
|
+
|
|
154
|
+
```bash
|
|
155
|
+
# Utiliser le template
|
|
156
|
+
cp templates/external-services/IService.cs src/Services/
|
|
157
|
+
```
|
|
158
|
+
|
|
159
|
+
### .NET Blazor
|
|
160
|
+
|
|
161
|
+
Templates pour projets .NET 8.0 Blazor Server :
|
|
162
|
+
|
|
163
|
+
- Structure de projet standard
|
|
164
|
+
- Configuration de base
|
|
165
|
+
- Exemples de composants
|
|
166
|
+
|
|
167
|
+
```bash
|
|
168
|
+
# Initialiser un projet Blazor
|
|
169
|
+
mdan init mon-projet --template dotnet-blazor
|
|
170
|
+
```
|
|
171
|
+
|
|
172
|
+
### SQL Server
|
|
173
|
+
|
|
174
|
+
Templates SQL Server pour bases de données :
|
|
175
|
+
|
|
176
|
+
- `schema.sql` — Schéma de base de données
|
|
177
|
+
- `stored-procedures.sql` — Procédures stockées
|
|
178
|
+
- `functions.sql` — Fonctions SQL
|
|
179
|
+
|
|
180
|
+
```bash
|
|
181
|
+
# Utiliser les templates SQL
|
|
182
|
+
cp templates/sql-server/schema.sql database/
|
|
183
|
+
```
|
|
184
|
+
|
|
185
|
+
---
|
|
186
|
+
|
|
132
187
|
## 🔧 Fonctionnalités Avancées
|
|
133
188
|
|
|
189
|
+
### MDAN-AUTO v1.0 — Mode Autonome
|
|
190
|
+
|
|
191
|
+
Développement entièrement autonome sans intervention humaine :
|
|
192
|
+
|
|
193
|
+
```bash
|
|
194
|
+
# Démarrer le mode autonome
|
|
195
|
+
mdan auto
|
|
196
|
+
|
|
197
|
+
# Reprendre depuis une sauvegarde
|
|
198
|
+
mdan resume /tmp/mdan-save-1234567890.json
|
|
199
|
+
```
|
|
200
|
+
|
|
201
|
+
**Phases Autonomes** :
|
|
202
|
+
1. **LOAD** — Chargement du contexte et des exigences
|
|
203
|
+
2. **DISCOVER** — Analyse des exigences et user stories
|
|
204
|
+
3. **PLAN** — Plan d'implémentation détaillé (#PHASE1, #PHASE2...)
|
|
205
|
+
4. **ARCHITECT** — Conception de l'architecture système
|
|
206
|
+
5. **IMPLEMENT** — Exécution de l'implémentation
|
|
207
|
+
6. **TEST** — Tests complets (unitaires, intégration, E2E)
|
|
208
|
+
7. **DEPLOY** — Déploiement en production
|
|
209
|
+
8. **DOC** — Génération de la documentation
|
|
210
|
+
|
|
211
|
+
**Fonctionnalités Clés** :
|
|
212
|
+
- 🔄 Sauvegarde automatique à 80% des tokens
|
|
213
|
+
- 🎯 Débats multi-agents pour décisions complexes
|
|
214
|
+
- ⚡ Exécution séquentielle sans pause
|
|
215
|
+
- 🚨 Fail-fast sur erreurs critiques
|
|
216
|
+
- 📊 Signaux de progression (PHASE X COMPLETE ✅)
|
|
217
|
+
|
|
134
218
|
### Scenarios (Better Agents)
|
|
135
219
|
|
|
136
220
|
Tests conversationnels end-to-end pour valider le comportement des agents.
|
|
@@ -187,8 +271,36 @@ mdan module add [nom] # Installer un module
|
|
|
187
271
|
mdan oc # Copier l'orchestrateur
|
|
188
272
|
mdan agent [nom] # Voir un agent
|
|
189
273
|
mdan skills # Lister les skills
|
|
274
|
+
mdan auto # Démarrer le mode autonome
|
|
275
|
+
mdan resume <save-file> # Reprendre depuis une sauvegarde
|
|
190
276
|
mdan mcp [action] # Config MCP
|
|
191
277
|
mdan prompt [action] # Gérer les prompts
|
|
278
|
+
mdan crewai <command> # CrewAI integration commands
|
|
279
|
+
```
|
|
280
|
+
|
|
281
|
+
### CrewAI CLI Commands
|
|
282
|
+
|
|
283
|
+
```bash
|
|
284
|
+
# Initialize CrewAI in a project
|
|
285
|
+
mdan crewai init
|
|
286
|
+
|
|
287
|
+
# Run autonomous mode
|
|
288
|
+
mdan crewai auto "Build a todo app"
|
|
289
|
+
|
|
290
|
+
# Start multi-agent debate
|
|
291
|
+
mdan crewai debate "Best architecture for a microservices app"
|
|
292
|
+
|
|
293
|
+
# Execute task with specific agent
|
|
294
|
+
mdan crewai agent dev "Implement user authentication"
|
|
295
|
+
|
|
296
|
+
# Execute specific flow
|
|
297
|
+
mdan crewai flow discovery "Analyze requirements"
|
|
298
|
+
|
|
299
|
+
# List available skills
|
|
300
|
+
mdan crewai skills
|
|
301
|
+
|
|
302
|
+
# Show CrewAI status
|
|
303
|
+
mdan crewai status
|
|
192
304
|
```
|
|
193
305
|
|
|
194
306
|
---
|
|
@@ -215,12 +327,31 @@ projet/
|
|
|
215
327
|
│ │ ├── test.md
|
|
216
328
|
│ │ └── ...
|
|
217
329
|
│ └── skills/ # Skills installés
|
|
330
|
+
├── integrations/
|
|
331
|
+
│ └── crewai/ # CrewAI integration
|
|
332
|
+
│ ├── agents/ # 8 CrewAI agents
|
|
333
|
+
│ ├── flows/ # 4 CrewAI flows
|
|
334
|
+
│ ├── tools/ # Custom tools (Serper, SQL, File)
|
|
335
|
+
│ ├── skills/ # Skill router
|
|
336
|
+
│ └── orchestrator.py # Main orchestrator
|
|
218
337
|
├── tests/
|
|
219
338
|
│ ├── scenarios/ # Tests conversationnels
|
|
220
339
|
│ └── evaluations/ # Évaluations benchmarks
|
|
221
340
|
├── templates/
|
|
222
|
-
│ ├──
|
|
223
|
-
│
|
|
341
|
+
│ ├── external-services/ # Templates services externes génériques
|
|
342
|
+
│ ├── dotnet-blazor/ # Templates .NET 8.0 Blazor Server
|
|
343
|
+
│ ├── sql-server/ # Templates SQL Server
|
|
344
|
+
│ ├── prompts/ # Prompts versionnés (YAML)
|
|
345
|
+
│ └── prompts.json # Registre des prompts
|
|
346
|
+
├── phases/
|
|
347
|
+
│ ├── auto-01-load.md # Phase LOAD (autonome)
|
|
348
|
+
│ ├── auto-02-discover.md # Phase DISCOVER (autonome)
|
|
349
|
+
│ ├── auto-03-plan.md # Phase PLAN (autonome)
|
|
350
|
+
│ ├── auto-04-architect.md # Phase ARCHITECT (autonome)
|
|
351
|
+
│ ├── auto-05-implement.md # Phase IMPLEMENT (autonome)
|
|
352
|
+
│ ├── auto-06-test.md # Phase TEST (autonome)
|
|
353
|
+
│ ├── auto-07-deploy.md # Phase DEPLOY (autonome)
|
|
354
|
+
│ └── auto-08-doc.md # Phase DOC (autonome)
|
|
224
355
|
├── mdan_output/ # Artifacts générés par les agents
|
|
225
356
|
├── .cursorrules # Pour Cursor
|
|
226
357
|
├── .windsurfrules # Pour Windsurf
|
|
@@ -228,6 +359,7 @@ projet/
|
|
|
228
359
|
├── .github/ # Pour Copilot
|
|
229
360
|
├── .mcp.json # Configuration MCP
|
|
230
361
|
├── AGENTS.md # Guidelines de développement
|
|
362
|
+
├── requirements_crewai.txt # CrewAI dependencies
|
|
231
363
|
└── MDAN-STATE.json # État de la session
|
|
232
364
|
```
|
|
233
365
|
|
|
@@ -250,6 +382,143 @@ MDAN se compose de plusieurs composants interconnectés :
|
|
|
250
382
|
|
|
251
383
|
---
|
|
252
384
|
|
|
385
|
+
## 🧪 Tests et Qualité
|
|
386
|
+
|
|
387
|
+
MDAN inclut une suite de tests complète pour garantir la qualité du code :
|
|
388
|
+
|
|
389
|
+
### Framework de Tests
|
|
390
|
+
|
|
391
|
+
- **pytest** — Framework de tests Python
|
|
392
|
+
- **pytest-cov** — Couverture de code
|
|
393
|
+
- **pytest-mock** — Mocking pour tests
|
|
394
|
+
|
|
395
|
+
### Types de Tests
|
|
396
|
+
|
|
397
|
+
| Type | Description | Emplacement |
|
|
398
|
+
|------|-------------|-------------|
|
|
399
|
+
| **Tests CLI** | Commandes CLI (init, attach, auto, etc.) | `tests/test_cli.py` |
|
|
400
|
+
| **Tests Templates** | Validation des templates | `tests/test_templates.py` |
|
|
401
|
+
| **Tests Services** | Patterns services externes | `tests/test_external_services.py` |
|
|
402
|
+
| **Tests Installation** | Vérification installation | `tests/test_installation.py` |
|
|
403
|
+
| **Tests Régression** | Pas de références wallet | `tests/test_regression.py` |
|
|
404
|
+
|
|
405
|
+
### Exécuter les Tests
|
|
406
|
+
|
|
407
|
+
```bash
|
|
408
|
+
# Tous les tests
|
|
409
|
+
pytest tests/ -v
|
|
410
|
+
|
|
411
|
+
# Avec couverture
|
|
412
|
+
pytest tests/ --cov=cli --cov-report=html
|
|
413
|
+
|
|
414
|
+
# Tests spécifiques
|
|
415
|
+
pytest tests/test_cli.py -v
|
|
416
|
+
```
|
|
417
|
+
|
|
418
|
+
### CI/CD
|
|
419
|
+
|
|
420
|
+
GitHub Actions configurés pour :
|
|
421
|
+
|
|
422
|
+
- **Tests automatiques** à chaque push
|
|
423
|
+
- **Linting** avec ruff et mypy
|
|
424
|
+
- **Couverture de code** minimale 80%
|
|
425
|
+
|
|
426
|
+
```bash
|
|
427
|
+
# Workflows CI/CD
|
|
428
|
+
.github/workflows/test.yml
|
|
429
|
+
.github/workflows/lint.yml
|
|
430
|
+
```
|
|
431
|
+
|
|
432
|
+
---
|
|
433
|
+
|
|
434
|
+
## 🤖 CrewAI Integration
|
|
435
|
+
|
|
436
|
+
MDANV2 includes a comprehensive CrewAI integration for intelligent multi-agent orchestration.
|
|
437
|
+
|
|
438
|
+
### Installation
|
|
439
|
+
|
|
440
|
+
```bash
|
|
441
|
+
# Note: CrewAI requires Python 3.10-3.13
|
|
442
|
+
pip install -r requirements_crewai.txt
|
|
443
|
+
```
|
|
444
|
+
|
|
445
|
+
### Environment Variables
|
|
446
|
+
|
|
447
|
+
```bash
|
|
448
|
+
export SERPER_API_KEY="your-serper-api-key"
|
|
449
|
+
export OPENAI_API_KEY="your-openai-api-key"
|
|
450
|
+
```
|
|
451
|
+
|
|
452
|
+
### Usage
|
|
453
|
+
|
|
454
|
+
```python
|
|
455
|
+
import asyncio
|
|
456
|
+
from integrations.crewai import CrewAIOrchestrator
|
|
457
|
+
|
|
458
|
+
async def main():
|
|
459
|
+
orchestrator = CrewAIOrchestrator(
|
|
460
|
+
project_path="/path/to/project",
|
|
461
|
+
llm=your_llm_instance,
|
|
462
|
+
auto_mode=True
|
|
463
|
+
)
|
|
464
|
+
|
|
465
|
+
# Run autonomous mode
|
|
466
|
+
result = await orchestrator.run_auto_mode("Build a todo app")
|
|
467
|
+
|
|
468
|
+
asyncio.run(main())
|
|
469
|
+
```
|
|
470
|
+
|
|
471
|
+
### Components
|
|
472
|
+
|
|
473
|
+
- **8 CrewAI Agents**: Product, Architect, UX, Dev, Test, Security, DevOps, Doc
|
|
474
|
+
- **4 CrewAI Flows**: Auto, Discovery, Build, Debate
|
|
475
|
+
- **3 Custom Tools**: Serper (web search), SQL (async database), File (file operations)
|
|
476
|
+
- **Skill Router**: Intelligent skill detection and routing (50+ skills)
|
|
477
|
+
|
|
478
|
+
### Example Scripts
|
|
479
|
+
|
|
480
|
+
See the `examples/` directory for complete examples:
|
|
481
|
+
|
|
482
|
+
- `examples/crewai_auto_mode.py` — Autonomous mode example
|
|
483
|
+
- `examples/crewai_debate.py` — Multi-agent debate example
|
|
484
|
+
- `examples/crewai_custom_crew.py` — Custom crew creation example
|
|
485
|
+
- `examples/crewai_with_sql.py` — SQL integration example (in-memory SQLite)
|
|
486
|
+
- `examples/crewai_with_serper.py` — Web search example
|
|
487
|
+
|
|
488
|
+
### Troubleshooting
|
|
489
|
+
|
|
490
|
+
**Python Version Issue**: CrewAI requires Python 3.10-3.13. If you're using Python 3.14, you'll need to use a different Python version.
|
|
491
|
+
|
|
492
|
+
```bash
|
|
493
|
+
# Check Python version
|
|
494
|
+
python --version
|
|
495
|
+
|
|
496
|
+
# Create a virtual environment with Python 3.13
|
|
497
|
+
python3.13 -m venv venv
|
|
498
|
+
source venv/bin/activate
|
|
499
|
+
pip install -r requirements_crewai.txt
|
|
500
|
+
```
|
|
501
|
+
|
|
502
|
+
**Import Errors**: Make sure you're running from the project root directory.
|
|
503
|
+
|
|
504
|
+
```bash
|
|
505
|
+
# Run from project root
|
|
506
|
+
cd /path/to/MDANV2
|
|
507
|
+
python examples/crewai_auto_mode.py
|
|
508
|
+
```
|
|
509
|
+
|
|
510
|
+
**Missing API Keys**: Ensure your environment variables are set.
|
|
511
|
+
|
|
512
|
+
```bash
|
|
513
|
+
# Check if keys are set
|
|
514
|
+
echo $OPENAI_API_KEY
|
|
515
|
+
echo $SERPER_API_KEY
|
|
516
|
+
```
|
|
517
|
+
|
|
518
|
+
### Documentation
|
|
519
|
+
|
|
520
|
+
See `core/crewai_orchestrator.md` for complete documentation.
|
|
521
|
+
|
|
253
522
|
## 📚 Documentation
|
|
254
523
|
|
|
255
524
|
| Document | Description |
|
|
@@ -257,6 +526,7 @@ MDAN se compose de plusieurs composants interconnectés :
|
|
|
257
526
|
| [ARCHITECTURE.md](ARCHITECTURE.md) | Architecture technique |
|
|
258
527
|
| [MDAN.md](MDAN.md) | Spécification complète |
|
|
259
528
|
| [AGENTS.md](AGENTS.md) | Guidelines de développement |
|
|
529
|
+
| [core/crewai_orchestrator.md](core/crewai_orchestrator.md) | CrewAI integration documentation |
|
|
260
530
|
| [docs/fr/](docs/fr/) | Documentation en français |
|
|
261
531
|
|
|
262
532
|
---
|