Can I vibecode Scholarcy?

Build a comprehensive personal substitute for Scholarcy, functioning as an advanced AI-powered academic research and summarization workspace.
Use exactly this stack: Next.js 15 + TypeScript + SQLite + TailwindCSS (for styling) + Playwright (for testing) + PDF.js (or similar for PDF parsing) + OpenAI SDK (or generic AI integration).

Primary job: Build a private application that allows users to import academic documents (PDF, Word, TXT) and automatically processes them into highly organized "Flashcards" containing structured summaries, key facts, methodologies, findings, and references.

Core Features to Implement:
1. Document Parsing & Import: A reliable document upload system that handles PDFs and extracts raw text cleanly.
2. Structured AI Summaries (Flashcards): Connect to an AI API to break down dense papers into a structured flashcard format. The flashcard must include: Title, Authors, Abstract, Key Findings, Methodology, and Conclusion.
3. Adaptive Reading Levels: Allow the user to toggle the summary complexity (e.g., Simple One-liners, Bulleted List, Researcher-Level Overview).
4. Auto-Highlighter (Robo-Highlighter™ clone): Automatically identify and highlight the most critical sentences within the extracted text.
5. Key Concepts & Spotlight: Extract and list key terminologies and their definitions from the text to help users grasp unfamiliar concepts quickly.
6. Library Management: Provide a searchable dashboard where users can save, tag, annotate, and organize their document summaries into collections or folders.
7. Bibliography Generation & Export: Generate citations for the document and allow users to export their structured summaries/flashcards to Markdown and CSV formats.

Technical constraints & Guidelines:
- Start from an empty folder and create the complete working project.
- Make the default mode single-user and private. All data (documents, summaries, tags) MUST be stored locally in SQLite.
- Put every secret and external credential (like the LLM API key) in .env and provide .env.example.
- Do not add analytics, telemetry, ads, or third-party accounts.
- Use realistic sample data that is clearly labelled and easy to delete to demonstrate the flashcard layout.
- Implement a highly polished, academic-focused UI. Use accessible keyboard navigation, labels, focus states, and sensible contrast.
- Include clear empty, loading (especially during AI processing), validation, success, and failure states.
- Where an external API is optional (or if the API key is missing), keep the app useful by allowing manual flashcard creation and explain the degraded mode.
- Write focused unit tests for the data model and the most important workflows (e.g., flashcard generation logic).
- Add one end-to-end smoke test using Playwright that proves the core loop works (uploading a mock document and viewing a summary).
- Create a detailed README with setup, architecture, data location, and limitations.
- Add scripts for install, development, test, build, and production start.