JSBrowser is a basic browser built as a Win10 JavaScript UWP app around the WebView HTML element. Its fun and relatively simple to implement tiny browser features in JavaScript and in this post I'm implementing crash resistance.

The normal DOM mechanisms for creating an HTML WebView create an in-process WebView, in which the WebView runs on a unique UI thread. But we can use the MSWebView constructor instead to create an out-of-process WebView in which the WebView runs in its own distinct WebView process. Unlike an in-process WebView, Web content running in an out-of-process WebView can only crash the WebView process and not the app process.

        this.replaceWebView = () => {
            let webview = document.querySelector("#WebView");
            // Cannot access webview.src - anything that would need to communicate with the webview process may fail
            let oldSrc = browser.currentUrl;
            const webviewParent = webview.parentElement;
            webviewParent.removeChild(webview);
            webview = new MSWebView();
            Object.assign(this, {
                "webview": webview
            });
            webview.setAttribute("id", "WebView");

            // During startup our currentUrl field is blank. If the WebView has crashed 
            // and we were on a URI then we may obtain it from this property.
            if (browser.currentUrl && browser.currentUrl != "") {
                this.trigger("newWebview");
                this.navigateTo(browser.currentUrl);
            }
            webviewParent.appendChild(webview);

I run replaceWebView during startup to replace the in-process WebView created via HTML markup with an out-of-process WebView. I could be doing more to dynamically copy styles, attributes, etc but I know what I need to set on the WebView and just do that.

When a WebView process crashes the corresponding WebView object is no longer useful and a new WebView element must be created. In fact if the old WebView object is used it may throw and will no longer have valid state. Accordingly when the WebView crashes I run replaceWebView again. Additionally, I need to store the last URI we've navigated to (browser.currentUrl in the above) since the crashed WebView object won't know what URI it is on after it crashes.

            webview.addEventListener("MSWebViewProcessExited", () => { 
                if (browser.currentUrl === browser.lastCrashUrl) {                   ++browser.lastCrashUrlCrashCount; 
                } 
                else { 
                    browser.lastCrashUrl = browser.currentUrl; 
                    browser.lastCrashUrlCrashCount = 1; 
                } 
                // If we crash again and again on the same URI, maybe stop trying to load that URI. 
                if (browser.lastCrashUrlCrashCount >= 3) { 
                    browser.lastCrashUrl = ""; 
                    browser.lastCrashUrlCrashCount = 0; 
                    browser.currentUrl = browser.startPage; 
                } 
                this.replaceWebView(); 
            }); 

I also keep track of the last URI that we recovered and how many times we've recovered that same URI. If the same URI crashes more than 3 times in a row then I assume that it will keep happening and I navigate to the start URI instead.

JavaScript Microsoft Store apps have some details related to activation that are specific to JavaScript Store apps and that are poorly documented which I’ll describe here.

StartPage syntax

The StartPage attributes in the AppxManifest.xml (Package/Applications/Application/@StartPage, Package/Applications/Extensions/Extension/@StartPage) define the HTML page entry point for that kind of activation. That is, Application/@StartPage defines the entry point for tile activation, Extension[@Category="windows.protocol"]/@StartPage defines the entry point for URI handling activation, etc. There are two kinds of supported values in StartPage attributes: relative Windows file paths and absolute URIs. If the attribute doesn’t parse as an absolute URI then it is instead interpreted as relative Windows file path.

This implies a few things that I’ll declare explicitly here. Windows file paths, unlike URIs, don’t have a query or fragment, so if you are using a relative Windows file path for your StartPage attribute you cannot include anything like ‘?param=value’ at the end. Absolute URIs use percent-encoding for reserved characters like ‘%’ and ‘#’. If you have a ‘#’ in your HTML filename then you need to percent-encode that ‘#’ for a URI and not for a relative Windows file path.

If you specify a relative Windows file path, it is turned into an ms-appx URI by changing all backslashes to forward slashes, percent-encoding reserved characters, and combining the result with a base URI of ms-appx:///. Accordingly the relative Windows file paths are relative to the root of your package. If you are using a relative Windows file path as your StartPage and need to switch to using a URI so you can include a query or fragment, you can follow the same steps above.

StartPage validity

The validity of the StartPage is not determined before activation. If the StartPage is a relative Windows file path for a file that doesn’t exist, or an absolute URI that is not in the Application Content URI Rules, or something that doesn’t parse as a Windows file path or URI, or otherwise an absolute URI that fails to resolve (404, bad hostname, etc etc) then the JavaScript app will navigate to the app’s navigation error page (perhaps more on that in a future blog post). Just to call it out explicitly because I have personally accidentally done this: StartPage URIs are not automatically included in the Application Content URI Rules and if you forget to include your StartPage in your ACUR you will always fail to navigate to that StartPage.

StartPage navigation

When your app is activated for a particular activation kind, the StartPage value from the entry in your app’s manifest that corresponds to that activation kind is used as the navigation target. If the app is not already running, the app is activated, navigated to that StartPage value and then the Windows.UI.WebUI.WebUIApplication activated event is fired (more details on the order of various events in a moment). If, however, your app is already running and an activation occurs, we navigate or don’t navigate to the corresponding StartPage depending on the current page of the app. Take the app’s current top level document’s URI and if after removing the fragment it already matches the StartPage value then we won’t navigate and will jump straight to firing the WebUIApplication activated event.

Since navigating the top-level document means destroying the current JavaScript engine instance and losing all your state, this behavior might be a problem for you. If so, you can use the MSApp.pageHandlesAllApplicationActivations(true) API to always skip navigating to the StartPage and instead always jump straight to firing the WebUIApplication activated event. This does require of course that all of your pages all handle all activation kinds about which any part of your app cares.

Application Content URI Rules (ACUR from now on) defines the bounds of the web that make up the Microsoft Store application. Package content via the ms-appx URI scheme is automatically considered part of the app. But if you have content on the web via http or https you can use ACUR to declare to Windows that those URIs are also part of your application. When your app navigates to URIs on the web those URIs will be matched against the ACUR to determine if they are part of your app or not. The documentation for how matching is done on the wildcard URIs in the ACUR Rule elements is not very helpful on MSDN so here are some notes.

Rules

You can have up to 100 Rule XML elements per ApplicationContentUriRules element. Each has a Match attribute that can be up to 2084 characters long. The content of the Match attribute is parsed with CreateUri and when matching against URIs on the web additional wildcard processing is performed. I’ll call the URI from the ACUR Rule the rule URI and the URI we compare it to found during app navigation the navigation URI.

The rule URI is matched to a navigation URI by URI component: scheme, username, password, host, port, path, query, and fragment. If a component does not exist on the rule URI then it matches any value of that component in the navigation URI. For example, a rule URI with no fragment will match a navigation URI with no fragment, with an empty string fragment, or a fragment with any value in it.

Asterisk

Each component except the port may have up to 8 asterisks. Two asterisks in a row counts as an escape and will match 1 literal asterisk. For scheme, username, password, query and fragment the asterisk matches whatever it can within the component.

Host

For the host, if the host consists of exactly one single asterisk then it matches anything. Otherwise an asterisk in a host only matches within its domain name label. For example, http://*.example.com will match http://a.example.com/ but not http://b.a.example.com/ or http://example.com/. And http://*/ will match http://example.com, http://a.example.com/, and http://b.a.example.com/. However the Store places restrictions on submitting apps that use the http://* rule or rules with an asterisk in the second effective domain name label. For example, http://*.com is also restricted for Store submission.

Path

For the path, an asterisk matches within the path segment. For example, http://example.com/a/*/c will match http://example.com/a/b/c and http://example.com/a//c but not http://example.com/a/b/b/c or http://example.com/a/c

Additionally for the path, if the path ends with a slash then it matches any path that starts with that same path. For example, http://example.com/a/ will match http://example.com/a/b and http://example.com/a/b/c/d/e/, but not http://example.com/b/.

If the path doesn’t end with a slash then there is no suffix matching performed. For example, http://example.com/a will match only http://example.com/a and no URIs with a different path.

As a part of parsing the rule URI and the navigation URI, CreateUri will perform URI normalization and so the hostname and scheme will be made lower case (casing matters in all other parts of the URI and case sensitive comparisons will be performed), IDN normalization will be performed, ‘.’ and ‘..’ path segments will be resolved and other normalizations as described in the CreateUri documentation.

• The location object's search, hash, and protocol properties are all misnomers that lead to confusion about the correct terms:
  • The 'search' property returns the URI's query property. The query property isn't limited to containing search terms.
  • The 'hash' property returns the URI's fragment property. This one is just named after its delimiter. It should be called the fragment.
  • The 'protocol' property returns the URI's scheme property. A URI's scheme isn't necessarily a protocol. The http URI scheme of course uses the HTTP protocol, but the https URI scheme is the HTTP protocol over SSL/TLS - there is no HTTPS protocol. Similarly for something like mailto - there is no mailto wire protocol.
• The 'hash' and 'search' location properties both return null in the case that their corresponding URI property doesn't exist or if its the empty string. A URI with no query property and a URI with an empty string query property that are otherwise the same, are not equal URIs and are allowed by HTTP to return different content. Similarly for the fragment. Unless the specific URI scheme defines otherwise, an empty query or hash isn't the same as no query or hash.

Level 4 and level 6 of the Stripe CTF had solutions around XSS.

Level 4

Code

> Registered Users 

<%@registered_users.each do |user| %>
<%last_active = user[:last_active].strftime('%H:%M:%S UTC') %>
<%if @trusts_me.include?(user[:username]) %>

• 
<%= user[:username] %>
(password: <%= user[:password] %>, last active <%= last_active %>)


Issue

The level 4 web application lets you transfer karma to another user and in doing so you are also forced to expose your password to that user. The main user page displays a list of users who have transfered karma to you along with their password. The password is not HTML encoded so we can inject HTML into that user's browser. For instance, we could create an account with the following HTML as the password which will result in XSS with that HTML:

This HTML runs script that uses jQuery to post to the transfer URI resulting in a transfer of karma from the attacked user to the attacker user, and also the attacked user's password.

Notes

Code review red flags in this case included lack of encoding when using user controlled content to create HTML content, storing passwords in plain text in the database, and displaying passwords generally. By design the web app shows users passwords which is a very bad idea.

Level 6

Code

...

    def self.safe_insert(table, key_values)
      key_values.each do |key, value|
        # Just in case people try to exfiltrate
        # level07-password-holder's password
        if value.kind_of?(String) &&
            (value.include?('"') || value.include?("'"))
          raise "Value has unsafe characters"
        end
      end

      conn[table].insert(key_values)
    end

Issue

This web app does a much better job than the level 4 app with HTML injection. They use encoding whenever creating HTML using user controlled data, however they don't use encoding when injecting JSON data into script (see post_data initialization above). This JSON data is the last five most recent messages sent on the app so we get to inject script directly. However, the system also ensures that no strings we write contains single or double quotes so we can't get out of the string in the JSON data directly. As it turns out, HTML lets you jump out of a script block using no matter where you are in script. For instance, in the middle of a value in some JSON data we can jump out of script. But we still want to run script, so we can jump right back in. So the frame so far for the message we're going to post is the following:

Stripe's web security CTF's Level 1 and level 2 of the Stripe CTF had issues with missing input validation solutions described below.

Level 1

Code

          $filename = 'secret-combination.txt';
      extract($_GET);
      if (isset($attempt)) {
        $combination = trim(file_get_contents($filename));
        if ($attempt === $combination) {

Issue

The issue here is the usage of the extract php method which extracts name value pairs from the map input parameter and creates corresponding local variables. However this code uses $_GET which contains a map of name value pairs passed in the query of the URI. The expected behavior is to get an attempt variable out, but since no input validation is done I can provide a filename variable and overwrite the value of $filename. Providing an empty string gives an empty string $combination which I can match with an empty string $attempt. So without knowing the combination I can get past the combination check.

Notes

Code review red flag in this case was the direct use of $_GET with no validation. Instead of using extract the developer could try to extract specifically the attempt variable manually without using extract.

Level 2

Code

    $dest_dir = "uploads/";
    $dest = $dest_dir . basename($_FILES["dispic"]["name"]);
    $src = $_FILES["dispic"]["tmp_name"];
    if (move_uploaded_file($src, $dest)) {
      $_SESSION["dispic_url"] = $dest;
      chmod($dest, 0644);
      echo "
Successfully uploaded your display picture.
";
    }

Issue

This code accepts POST uploads of images but with no validation to ensure it is not an arbitrary file. And even though it uses chmod to ensure the file is not executable, things like PHP don't require a file to be executable in order to run them. Accordingly, one can upload a PHP script, then navigate to that script to run it. My PHP script dumped out the contents of the file we're interested in for this level:

Notes

Code review red flags include manual file management, chmod, and use of file and filename inputs without any kind of validation. If this code controlled the filename and ensured that the extension was one of a set of image extensions, this would solve this issue. Due to browser mime sniffing its additionally a good idea to serve a content-type that starts with "image/" for these uploads to ensure browsers treat these as images and not sniff for script or HTML.

Stripe's web security CTF's level 0 and level 3 had SQL injection solutions described below.

Level 0

Code

app.get('/*', function(req, res) {
  var namespace = req.param('namespace');

  if (namespace) {
    var query = 'SELECT * FROM secrets WHERE key LIKE ? || ".%"';
    db.all(query, namespace, function(err, secrets) {

Issue

There's no input validation on the namespace parameter and it is injected into the SQL query with no encoding applied. This means you can use the '%' character as the namespace which is the wildcard character matching all secrets.

Notes

Code review red flag was using strings to query the database. Additional levels made this harder to exploit by using an API with objects to construct a query rather than strings and by running a query that only returned a single row, only ran a single command, and didn't just dump out the results of the query to the caller.

Level 3

Code

@app.route('/login', methods=['POST'])
def login():
    username = flask.request.form.get('username')
    password = flask.request.form.get('password')

    if not username:
        return "Must provide username\n"

    if not password:
        return "Must provide password\n"

    conn = sqlite3.connect(os.path.join(data_dir, 'users.db'))
    cursor = conn.cursor()

    query = """SELECT id, password_hash, salt FROM users
               WHERE username = '{0}' LIMIT 1""".format(username)
    cursor.execute(query)

    res = cursor.fetchone()
    if not res:
        return "There's no such user {0}!\n".format(username)
    user_id, password_hash, salt = res

    calculated_hash = hashlib.sha256(password + salt)
    if calculated_hash.hexdigest() != password_hash:
        return "That's not the password for {0}!\n".format(username)

Issue

There's little input validation on username before it is used to constrcut a SQL query. There's no encoding applied when constructing the SQL query string which is used to, given a username, produce the hashed password and the associated salt. Accordingly one can make username a part of a SQL query command which ensures the original select returns nothing and provide a new SELECT via a UNION that returns some literal values for the hash and salt. For instance the following in blue is the query template and the red is the username injected SQL code:

SELECT id, password_hash, salt FROM users WHERE username = 'doesntexist' UNION SELECT id, ('5e884898da28047151d0e56f8dc6292773603d0d6aabbdd62a11ef721d1542d8') AS password_hash, ('word') AS salt FROM users WHERE username = 'bob' LIMIT 1

Notes

Code review red flag is again using strings to query the database. Although this level was made more difficult by using an API that returns only a single row and by using the execute method which only runs one command. I was forced to (as a SQL noob) learn the syntax of SELECT in order to figure out UNION and how to return my own literal values.