https://github.com/shivamr021/codebase-intelligence-web
IntelliRepo — an AI-powered repository intelligence platform with architecture visualization, dependency analysis, bug detection, and natural language codebase querying.
https://github.com/shivamr021/codebase-intelligence-web
architecture-diagram dependency-graph llm rag tanstack webapp
Last synced: 9 days ago
JSON representation
IntelliRepo — an AI-powered repository intelligence platform with architecture visualization, dependency analysis, bug detection, and natural language codebase querying.
- Host: GitHub
- URL: https://github.com/shivamr021/codebase-intelligence-web
- Owner: shivamr021
- License: mit
- Created: 2026-05-28T07:28:51.000Z (29 days ago)
- Default Branch: master
- Last Pushed: 2026-06-05T19:03:45.000Z (21 days ago)
- Last Synced: 2026-06-05T20:23:11.371Z (21 days ago)
- Topics: architecture-diagram, dependency-graph, llm, rag, tanstack, webapp
- Language: TypeScript
- Homepage: https://codebase-intelligence.shivamrathod145.workers.dev/
- Size: 195 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Codebase Intelligence Web
🔗 Live Demo: https://codebase-intelligence.shivamrathod145.workers.dev/
Frontend application for Codebase Intelligence — an AI-powered repository intelligence platform that helps developers understand unfamiliar codebases through semantic search, dependency graph visualization, architecture analysis, and AI-assisted code review.
## Overview
Understanding a large repository often requires manually exploring files, tracing imports, and identifying architectural boundaries before meaningful contributions can be made.
Codebase Intelligence Web provides a unified interface for interacting with the Codebase Intelligence backend and exploring repository insights through an intuitive web experience.
The application enables developers to:
* Index public GitHub repositories
* Ask natural-language questions about codebases
* Visualize dependency graphs
* Generate architecture summaries and Mermaid diagrams
* Review AI-generated bug reports
* Explore repository metadata through a single interface
---
## Features
### Repository Indexing
Submit a public GitHub repository URL and trigger repository ingestion through the backend analysis pipeline.
The application automatically:
* Validates repository URLs
* Checks whether a repository has already been indexed
* Displays ingestion results and repository metadata
### Semantic Repository Q&A
Ask questions such as:
* "How does authentication work?"
* "Where is the dependency graph generated?"
* "How is metadata stored?"
The frontend displays AI-generated answers retrieved from indexed repository knowledge.
### Dependency Graph Visualization
Explore repository relationships through interactive dependency graph visualizations.
Provides access to:
* Repository graph statistics
* Dependency relationships
* Interactive graph rendering
### Architecture Analysis
Generate:
* High-level architecture summaries
* Mermaid architecture diagrams
Architecture information is retrieved from backend repository analysis services.
### AI-Assisted Bug Reports
View structured code review findings generated by the backend.
Reports may include:
* Logic issues
* Crash risks
* Security concerns
* Incorrect API usage
* Dependency-related issues
---
## Example Workflow
### 1. Enter Repository URL
Provide a public GitHub repository URL.
Example:
```text
https://github.com/user/repository
```
### 2. Index Repository
The application:
* Validates the repository
* Checks existing indexed data
* Starts ingestion when necessary
### 3. Explore Repository Insights
After indexing, access:
* Q&A
* Architecture
* Dependency Graph
* Bug Reports
from the application interface.
---
## Architecture
### Core UI Modules
#### Repository Ingestion
Responsible for:
* URL validation
* Repository indexing requests
* Repository status checks
#### Q&A Interface
Responsible for:
* User questions
* Displaying AI-generated responses
* Repository knowledge exploration
#### Architecture Viewer
Responsible for:
* Architecture summaries
* Mermaid diagram rendering
#### Graph Viewer
Responsible for:
* Dependency graph visualization
* Repository graph statistics
#### Bug Report Viewer
Responsible for:
* Displaying AI-generated review findings
* Organizing structured issue reports
---
## Tech Stack
### Frontend
* React 19
* TypeScript
* Vite
### Routing & State Management
* TanStack Router
* TanStack Query
### UI
* Tailwind CSS v4
* shadcn/ui
* Radix UI
* Lucide React
### Visualization
* Mermaid
### Deployment
* Cloudflare Workers
---
## Project Structure
```text
src/
├── components/
│ ├── ArchitectureTab.tsx
│ ├── BugReportTab.tsx
│ ├── GraphTab.tsx
│ ├── MermaidDiagram.tsx
│ └── QATab.tsx
│
├── lib/
│ └── api.ts
│
├── routes/
│ ├── __root.tsx
│ └── index.tsx
│
└── styles.css
```
---
## Installation
### Clone Repository
```bash
git clone https://github.com/shivamr021/codebase-intelligence-web.git
cd codebase-intelligence-web
```
### Install Dependencies
```bash
npm install
```
### Environment Variables
Create a `.env` file:
```env
VITE_API_BASE_URL=http://localhost:8000
```
---
## Run Locally
```bash
npm run dev
```
Application:
```text
http://localhost:3000
```
---
## Backend
The frontend communicates with the Codebase Intelligence backend API for:
* Repository ingestion
* Semantic search
* Architecture generation
* Dependency graph generation
* Bug review
Backend Repository:
https://github.com/shivamr021/codebase-intelligence
---
## Limitations
* Repository analysis depends on backend processing.
* Architecture diagrams represent repository structure and may not fully reflect runtime behavior.
* AI-generated bug reports should be reviewed manually.
* Large repositories may require longer indexing times.
---
## AI-Assisted Development
This project was developed with significant AI assistance during:
* UI design iterations
* Component development
* State management implementation
* Debugging workflows
* Documentation generation
All final implementation decisions, integration work, testing, and project direction were performed by the author.
AI was used as a development assistant rather than an autonomous code generator.
---
## Future Improvements
* Repository history exploration
* Repository comparison workflows
* Improved graph filtering
* Enhanced architecture visualizations
* Multi-repository analysis
* User authentication and saved workspaces
---
## Author
**Shivam Rathod**
GitHub:
https://github.com/shivamr021
LinkedIn:
https://www.linkedin.com/in/shivamrathod021/
---
## License
This project is licensed under the MIT License.