fake-app-scraper / README.md
sylvain471's picture
Update README.md
fe5c202 verified
---
title: Fake App Scraper
emoji: πŸ“ˆ
colorFrom: green
colorTo: pink
sdk: docker
app_file: app.py
pinned: false
license: mit
short_description: uses playwright to scrape a fake app hosted on vercel
tags: ["mcp-server-track"]
---
## EDIT: post deadline modification
following a discussion with **abubakar** on discord I added a feature that check whether a token is given by a user (UI version) or via request header (mcp version). This allows to experiment in the UI whereas the pre-deadline version only allowed for mcp use as was describe in the pre-deadline README.
Also README was edited after deadline to fix the tag->tags issue.
The last commit before deadline was a199b381ac05f2e732e3de120e24711567949a3a
## Fake app scraper
link to test video with Claude https://youtu.be/WhLegRIdpRo
<div style="font-family: sans-serif; line-height: 1.6;">
<p>
This app uses Playwright to log in and scrape the content of the dashboard of the fake app
<a href="https://fake-app-omega.vercel.app" target="_blank">fake-app-omega.vercel.app</a>.
</p>
<p>
The url, username and password to log in to the fake-app dashboard are embedded in the token provided by the user for UI use or passed to the mcp-server via a token in the request header.
</p>
<p>
You can test the mcp-server tools in Claude desktop using the following mcp-server config
</p>
</div>
Claude config
```json
{
"mcpServers": {
"gradio": {
"command": "npx",
"args": [
"mcp-remote",
"https://agents-mcp-hackathon-fake-app-scraper.hf.space/gradio_api/mcp/sse",
"--header",
"token:aHR0cHM6Ly9mYWtlLWFwcC1vbWVnYS52ZXJjZWwuYXBwL2xvZ2luPHNlcD5ncmFkaW88c2VwPm1jcA==",
],
}
}
}
```
<div style="font-family: sans-serif; line-height: 1.6;">
<p>Fake App</p>
<img src="login.png" alt="login" style="max-width: 35%; margin-right: 10px;" />
<img src="dashboard.png" alt="dashboard" style="max-width: 35%;" />
</div>