AIForge
AIForge è un sistema bidirezionale per trasformare output disordinati di AI (ChatGPT, Claude, Mistral) in progetti software strutturati e mantenibili, e viceversa. Supporta Flutter, Python, Node.js, TypeScript, Java, C#, con gestione avanzata dei conflitti, ottimizzazione per limiti di token, e integrazione con Git e modelli locali.
🚀 Installazione
Base (solo CLI)
pip install aiforge
Con supporto per modelli locali
pip install aiforge[local-ai]
Per sviluppo
git clone https://github.com/tuo-username/AIForge.git
cd AIForge
pip install -e .[dev,local-ai]
🛠 Funzionalità
| Comando | Descrizione | Esempio |
|---|---|---|
ai2project |
Da dump AI a progetto strutturato | aiforge ai2project dump.txt my_project --mode clean |
project2ai |
Da progetto a file AI-friendly | aiforge project2ai my_project export.md --max-tokens 4000 |
sync |
Sincronizza progetto con nuovo output AI | aiforge sync new_dump.txt my_project --mode patch --non-interactive |
validate |
Valida la struttura del progetto | aiforge validate my_project |
init-git |
Configura hooks Git per validazione automatica | aiforge init-git my_project |
refine |
Migliora un progetto con AI locale | aiforge refine my_project --model mistralai/Mistral-7B-Instruct-v0.1 |
📦 Esempi d'Uso
1. Da AI a Progetto
aiforge ai2project dump.txt my_flutter_project --mode clean
2. Sincronizzazione con Gestione Conflitti
aiforge sync new_dump.txt my_project --mode patch --non-interactive
3. Raffinamento con AI Locale
aiforge refine my_project --model mistralai/Mistral-7B-Instruct-v0.1
🔧 Integrazione con Git
AIForge include pre-commit hooks per validare automaticamente il progetto prima di ogni commit:
aiforge init-git my_project
📂 Struttura di un Progetto AIForge
my_project/
├── lib/
│ ├── main.dart
│ └── utils.dart
├── pubspec.yaml
└── .git/
└── hooks/
└── pre-commit # Hook automatico per validazione
🤖 Supporto per Modelli Locali
AIForge supporta modelli come:
- Mistral-7B
- Llama2-70B
- CodeLlama
Esempio di raffinamento con modello locale:
aiforge refine my_project --model mistralai/Mistral-7B-Instruct-v0.1
🧪 Test
Esegui i test con:
pytest tests/
🤝 Contribuire
- Forka il repository.
- Crea un branch:
git checkout -b feature/nome-feature. - Committa le modifiche:
git commit -m "Aggiunta feature". - Pusha il branch:
git push origin feature/nome-feature. - Apri una Pull Request.
📄 Licenza
MIT
Metadata
Release files for aiforge-manager 1.0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| aiforge_manager-1.0.2.tar.gz | 8.7 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| aiforge_manager-1.0.2-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 20.0 kB
Release files / aiforge_manager-1.0.2.tar.gz
| Download URL | aiforge_manager-1.0.2.tar.gz |
|---|---|
| Size | 8.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
6d7b3c08da30f0b2e9dbed4bca7a1c6476045a2e997d232df7cf6da15e1eaf57
|
|
BLAKE2b-256 checksum How to use checksums |
1e8e6a6948a1f90165ec22c8f2e5a02be632edcaff0366f3ca175a5916b2e736
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.3
|
Release files / aiforge_manager-1.0.2-py3-none-any.whl
| Download URL | aiforge_manager-1.0.2-py3-none-any.whl |
|---|---|
| Size | 11.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
bb0fc4417496d0032e65277327fbfaefdd1af1b5f729e8f2fad9d4984c9db5d1
|
|
BLAKE2b-256 checksum How to use checksums |
f0b94bfcb379e03bcc77af24a4b5000dbd20daa3910200f65f3d5eabcfa49a7f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.3
|
Security for aiforge-manager 1.0.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.