https://github.com/florianibach/github-repo-raw-urls-ui
GitHub Raw URL Explorer is a small web UI that scans a public GitHub repository and lists raw.githubusercontent.com URLs for all files in the selected branch.
https://github.com/florianibach/github-repo-raw-urls-ui
ai ai-tools code-explorer developer-tools docker docker-compose github github-api llm llms promt-engineering raw-github repository-browser self-hosted selfhost web-ui
Last synced: 5 months ago
JSON representation
GitHub Raw URL Explorer is a small web UI that scans a public GitHub repository and lists raw.githubusercontent.com URLs for all files in the selected branch.
- Host: GitHub
- URL: https://github.com/florianibach/github-repo-raw-urls-ui
- Owner: florianibach
- License: mit
- Created: 2025-12-29T13:48:40.000Z (5 months ago)
- Default Branch: master
- Last Pushed: 2025-12-29T21:33:53.000Z (5 months ago)
- Last Synced: 2026-01-01T20:58:59.117Z (5 months ago)
- Topics: ai, ai-tools, code-explorer, developer-tools, docker, docker-compose, github, github-api, llm, llms, promt-engineering, raw-github, repository-browser, self-hosted, selfhost, web-ui
- Language: Go
- Homepage:
- Size: 509 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# GitHub Raw URL Explorer
[](https://github.com/florianibach/github-repo-raw-urls-ui)
[](https://hub.docker.com/r/floibach/github-repo-raw-urls-ui)
**GitHub Raw URL Explorer** is a small web UI that scans a public GitHub repository and lists **raw.githubusercontent.com URLs** for all files in the selected branch.
---
This project is built and maintained in my free time.
If it helps you or saves you some time, you can support my work on [](https://buymeacoffee.com/floibach)
Thank you for your support!
## What is this?
The main idea behind this tool is to make it easier to work with **AI chatbots / LLMs**.
When chatting with AI systems, you usually cannot paste an entire repository because it is not allowed to crawl the whole repository recursively and posting each file is:
- too large
- too noisy
- easy to miss files
However, most AI tools *can* fetch and read **raw GitHub file URLs**.
This service helps you quickly generate a **complete list of raw file links**, so you can paste them into an AI chat and let the model:
- write or improve tests
- review or polish UI code
- generate documentation
- refactor or explain code
- reason about the whole repository structure
In short:
**Give the AI the repo via links instead of copy & paste.**
## Screenshots
### Repository scan
## Features
- Scan any **public GitHub repository**
- Select branch (default branch is detected automatically)
- Recursively lists all files
- Generates `raw.githubusercontent.com` URLs
- Copy all links with one click
- No authentication required
- No GitHub token needed
- Lightweight and self-contained
---
## Usage
1. Enter a public GitHub repository URL
Example:
```
https://github.com/florianibach/github-repo-raw-urls-ui
````
2. Optionally select a branch
(the default branch is marked)
3. Click **Scan**
4. Copy the generated raw URLs and paste them into your AI chat
---
## Docker
### docker run
```bash
docker run --rm --name github-repo-raw-urls-ui -p 8080:8080 floibach/github-repo-raw-urls-ui
````
Open in your browser:
```
http://localhost:8080
```
---
### docker compose
```yaml
version: "3.8"
services:
github-repo-raw-urls-ui:
image: floibach/github-repo-raw-urls-ui
container_name: github-repo-raw-urls-ui
ports:
- "8080:8080"
restart: unless-stopped
```
Start the service:
```bash
docker compose up -d
```
---
## Technical notes
* Written in Go
* Uses the GitHub REST API (unauthenticated)
* No data is stored
* No analytics
* No tracking
* All processing happens at request time
---
## Limitations
* Only public repositories are supported
* GitHub API rate limits apply (unauthenticated requests)
* Very large repositories may take longer to scan
---
## Disclaimer
This tool was developed in close collaboration with an AI chat assistant and refined iteratively through human–AI interaction.
The final design decisions, implementation, and maintenance remain entirely human-driven.
---
## License
MIT License