B3 is your personal data assistant designed to help you conquer bureaucracy and be free.
Note: This project is currently in the conceptual phase. The features and functionality described below represent the vision for B3. No code has been implemented yet.
B3 is envisioned as a personal AI assistant that acts as a dedicated guardian and servant for your sensitive documents and personal data. It lives within your own Google Drive, transforming a simple folder of documents (like IDs, passports, bills, and contracts) into a dynamic, intelligent, and queryable database of your life.
The goal is to profoundly reduce the mental, temporal, and emotional burden of administrative tasks, turning reactive frustration into proactive control.
The value of B3 is built on a virtuous cycle that empowers you to break through bureaucratic barriers with increasing ease and confidence over time.
You shouldn't have to work to be organized. B3 does the heavy lifting for you.
- Zero-Effort Organization: Simply drop your documents into a folder. B3's AI reads, classifies, and extracts key information automatically.
- Conversational Interface: Interact using simple, natural language. Ask "I need my documents to apply for a rental" and B3 will understand.
B3's core mission is to deliver exactly what you need, ensuring you get things right the first time.
- Goal-Oriented Retrieval: B3 doesn't just find files; it understands objectives. It provides a complete and accurate package of documents for your specific task.
- Prevents Human Error: By automatically reading details like expiration dates, B3 helps you avoid the common pitfalls of submitting incorrect or outdated information.
B3 becomes more valuable with every document you add and every interaction you have.
- A Living Database: As you add new documents, B3's knowledge base grows, making it progressively more comprehensive and capable.
- Reinforced Through Use: The more you use B3, the better it understands your data, creating a powerful positive feedback loop where its value compounds over time.
The planned architecture is simple and user-centric:
- Store: You place your personal documents (PDFs, images of IDs, etc.) into a designated
B3folder in your Google Drive. - Index: B3's backend services securely scan new documents, using AI (OCR and NLP) to extract key data (e.g., name, document number, expiry date) and automatically write this structured information into the document's "description" field in Google Drive. Your data stays with your document.
- Query: You ask B3 for what you need via a simple interface (e.g., a web app or command line). For example: "Gather my proof of identity and address for a new bank account."
- Gather: B3 intelligently queries the description fields of your documents and instantly assembles the required files into a practical format (e.g., a downloadable
.zipfolder, a temporary secure webpage with links, or a markdown summary).
This project is in its earliest stage. The core concepts and value proposition are defined here, but the architecture is not yet finalized and no code has been written.
This is the perfect time to get involved! All contributions, from high-level architectural ideas to feature suggestions, are welcome. Please feel free to open an issue to start a discussion.
This project will be licensed under the MIT License.
This project will be coded in Go, and will make heavy use of Gemini for everything, including generating this doc).