🧠 TaskRebuild

TaskRebuild is an intelligent, task-aware extension project designed to act as your automatic "Second Brain." By combining a React-powered Chrome Extension with a massively scalable Python/PostgreSQL backend, it creates a next-generation, context-aware productivity workspace.

---

🤔 What actually is TaskRebuild?

The Problem: Context Decay

When you research a complex topic or write code, you usually open 50 StackOverflow tabs, save a few disjointed bookmarks, and scribble some notes. If you close your browser and come back a week later, your "context" is entirely gone. You have no idea why you saved a specific link or how it connects to the overall problem you were trying to solve.

The Solution: An Automatic Second Brain

TaskRebuild completely solves this by tracking your research automatically and mapping it into a visual, searchable graph.

Whenever you browse the web, the TaskRebuild Chrome Extension silently observes your active tabs and ties them to a specific Task (e.g., "Fix Authentication Bug"). Behind the scenes, the Python backend uses Google's Gemini AI to read what you are looking at, understand the content, and map out exactly how your web searches connect to your final solutions.

When you need to find something months later, you don't just search for a keyword—you talk to a 3072-dimension semantic search engine that instantly pulls up the exact paragraph you read 6 months ago, highlighting the exact sentence you need.

---

✨ Core Technical Engines

🔍 1. Google-Clone RRF Search Engine

TaskRebuild doesn't just do basic text search. It utilizes a custom Reciprocal Rank Fusion (RRF) engine built directly into PostgreSQL that simultaneously calculates and merges:

• Semantic Meaning (3072-Dimension Vector embeddings powered by Google Gemini AI)
• Keyword Matches (tsvector Full-Text Search)
• Personal PageRank (Logarithmic algorithmic boosts for pages you visit most frequently)
• Recency Bias (Decay mathematical boosts for fresher resources)

The frontend features lightning-fast pg_trgm Fuzzy Autocomplete and HTML Snippet Highlighting to show you exactly why a resource matched your query.

🕸️ 2. Automated Knowledge Graph

Every time you switch contexts, open a tab, or save a resource, the backend records it. Using Gemini AI, TaskRebuild analyzes your unstructured activity and automatically builds a mathematical Graph (nodes and edges) mapping exactly how your research topics connect to each other.

🛡️ 3. Strict Context Boundaries

Unlike standard bookmark managers that dump everything into one global bucket, TaskRebuild strictly binds your resources and searches to specific Tasks. If you search for a bug while inside your "Cooking Project," the database mathematically walls off the search space so you don't get polluted results from your "Coding Project."

📈 4. Zero-Latency System Metrics

The backend features an asynchronous background monitor that continually tracks CPU, RAM, Python Runtime states, and Vector Database capacities, broadcasting them live to the Extension's beautiful UI Dashboard.

---

🏗️ Architecture Stack

• Frontend Chrome Extension: React 18, TypeScript, Vite, Manifest V3
• Backend API: FastAPI (Python), Pydantic, httpx
• Database: Supabase (PostgreSQL), pgvector (HNSW/IVFFlat), pg_trgm (Fuzzy Logic)
• AI Provider: Google Gemini API (embedding-001 & Text Models)
• API Documentation: Integrated Scalar UI (/docs)

---

🚀 Installation & Setup

1. The Database (Supabase)

1. Create a new Supabase project.
2. Enable the pgvector and pg_trgm extensions.
3. Run the provided supabase_schema.sql in the SQL Editor to generate the tables, Row Level Security (RLS) policies, and the complex RRF Search RPCs.

2. The Python Backend

cd render-backend
python -m venv venv
source venv/bin/activate  # (or venv\Scripts\activate on Windows)
pip install -r requirements.txt

Create a .env file in the render-backend directory:

SUPABASE_URL=your_supabase_url
SUPABASE_SERVICE_ROLE_KEY=your_supabase_service_role_key
GEMINI_API_KEY=your_google_gemini_api_key
GRAPH_AI_PROVIDER=gemini
ALLOWED_ORIGINS=chrome-extension://<your-extension-id>

Start the server:

uvicorn app.main:app --reload

Navigate to http://localhost:8000/docs to view the beautiful Scalar API Documentation!

3. The Chrome Extension

cd taskrebuild/extension
npm install
npm run build

1. Open Google Chrome and go to chrome://extensions/
2. Enable Developer Mode in the top right.
3. Click Load Unpacked and select the taskrebuild/extension/dist directory.

---

🗺️ Roadmap & Future Upgrades

• Autonomous Web Scraper: A background worker that downloads, strips, and embeds the full HTML body of saved URLs for deep-quote searching.
• Obsidian-Style Graph UI: A 3D force-directed React graph to visually explore your Task nodes.
• GraphRAG Chatbot: A conversational UI that traverses the Knowledge Graph to answer complex questions about your saved resources.