How arnav-aka-guy/repository-scanner-engine works
Repository Scanner Engine is a codebase-intelligence application: users scan local or public GitHub repositories, then explore parsed source, dependency structure, search code, ask citation-grounded questions, and review health and technical debt. The backend coordinates scanning, graph analysis, embeddings, retrieval, and LLM responses; the frontend exposes these workflows. The README documents several capabilities whose implementation files were not sampled, so those connections are shown only where documented; unsampled internal wiring is omitted or marked unverified.
Main entry points: main.py FastAPI runtime · App.tsx React application · repository.py Repository scanning API · service.py Incremental parser
Components 28
Desktop and UI
- Electron shell
desktop/main.js - React application
frontend/src/App.tsx - Repository explorer
frontend/src/pages/RepositoryExplorer.tsx - Architecture viewer
frontend/src/pages/ArchitectureViewer.tsx - Workspace state
frontend/src/stores/workspaceStore.ts - Health dashboard
frontend/src/pages/HealthDashboard.tsx
API workflows
- Repository scanning API
backend/api/repository.py - Graph and visualization API
backend/api/visualization.py - Chat API
backend/api/chat.py - Health and debt APIs
backend/api/health_score.py
Repository analysis
- Incremental parser
backend/parser/service.py - Repository scanner
backend/parser/scanner.py - Language parsers
backend/parser/python_parser.py - Graph analysis
backend/graph/service.py - Health and debt analysis
backend/services/health_score.py - GitHub clone service
backend/services/github_scanner.py
Search and AI
- Embedding index
backend/embeddings/service.py - Per-repository vector store
backend/embeddings/store.py - Hybrid retrieval
backend/retrieval/service.py - LLM orchestration
backend/llm/service.py - LLM providers
backend/llm/provider_manager.py
Runtime platform
- FastAPI runtime
backend/main.py - Request security
backend/security/auth.py - Optional database
backend/db/database.py
Other components
- Engineer
- Code repository
- GitHub
- Configured LLM
Connections 33
- Engineer → Electron shell: uses
- Electron shell → React application: hosts
- Engineer → React application: uses
- React application → Repository explorer: routes to
frontend/src/App.tsx - React application → Architecture viewer: routes to
frontend/src/App.tsx - Repository explorer → Workspace state: uses
frontend/src/pages/RepositoryExplorer.tsx - Architecture viewer → Workspace state: uses
frontend/src/pages/ArchitectureViewer.tsx - Engineer → Health dashboard: reviews
- React application → Health dashboard: routes to
frontend/src/App.tsx - React application → FastAPI runtime: HTTP requests
- FastAPI runtime → Repository scanning API: mounts
backend/main.py - FastAPI runtime → Graph and visualization API: mounts
backend/main.py - FastAPI runtime → Chat API: mounts
backend/main.py - FastAPI runtime → Request security: authenticates
backend/main.py - FastAPI runtime → Optional database: initializes
backend/main.py - Code repository → Repository scanning API: scanned by
backend/api/repository.py - Repository scanning API → Incremental parser: parses
backend/api/repository.py - Repository scanning API → Embedding index: indexes
backend/api/repository.py - Repository scanning API → GitHub clone service: clones via
backend/api/repository.py - GitHub → GitHub clone service: clones from
- Incremental parser → Repository scanner: discovers files
backend/parser/service.py - Incremental parser → Language parsers: parses files
backend/parser/service.py - Graph and visualization API → Graph analysis: requests graphs
backend/api/visualization.py - Graph analysis → Incremental parser: uses parsed files
- Embedding index → Per-repository vector store: stores vectors
backend/embeddings/service.py - Chat API → Hybrid retrieval: retrieves context
- Chat API → LLM orchestration: requests answers
- Hybrid retrieval → Embedding index: searches vectors
backend/retrieval/service.py - Hybrid retrieval → Graph analysis: retrieves structure
backend/retrieval/service.py - LLM orchestration → LLM providers: generates through
backend/llm/service.py - LLM providers → Configured LLM: queries provider
- Health dashboard → Health and debt APIs: requests analysis
- Health and debt APIs → Health and debt analysis: runs analysis