นโยบายความปลอดภัยของเนื้อหา (CSP) สำหรับแบ็กเอนด์

Welcome to Content Security Policy (CSP) for Backend! You might think CSP is just for frontend developers, but the backend plays a crucial role.

CSP is a security standard that helps prevent client-side attacks like Cross-Site Scripting (XSS). It does this by telling the browser which resources (scripts, styles, images) it's allowed to load and execute.

The backend is responsible for delivering these rules to the browser.

The backend delivers CSP rules to the browser using a special HTTP response header called Content-Security-Policy. When the browser receives this header, it enforces the rules defined within it.

This means your backend application directly controls the security policy for your frontend. Let's see how a backend might set this header.

In a backend application, you'd typically add the CSP header to your HTTP response. This example simulates a backend sending a basic CSP header.

The default-src 'self' directive allows resources only from the same origin as the document.

As a backend developer, you'll often define directives that control various resource types. Some common ones include:

• script-src: Specifies valid sources for JavaScript.
• style-src: Specifies valid sources for stylesheets.
• img-src: Specifies valid sources for images.
• connect-src: Restricts URLs that can be loaded using scripting interfaces (e.g., AJAX, WebSockets).

Each directive can have multiple allowed sources, like 'self', https://example.com, or 'unsafe-inline' (generally to be avoided).

One common XSS attack vector is injecting inline scripts. CSP can block these, but sometimes inline scripts are necessary.

The backend can generate a unique, cryptographically secure nonce (Number Used Once) for each request. This nonce is added to both the CSP header and the allowed inline <script> tags.

Here's how a backend can generate a nonce. This nonce is then included in the Content-Security-Policy header (e.g., script-src 'nonce-YOUR_NONCE_HERE') and rendered into the HTML script tag.

The browser will only execute inline scripts that have a matching nonce attribute.

CSP isn't just about blocking; it's also about visibility. The backend can specify a reporting endpoint using the report-to directive (or older report-uri).

If a browser violates the CSP (e.g., tries to load a script from an unauthorized source), it will send a JSON report to this backend endpoint. Your backend can then log and analyze these reports to detect potential attacks or policy misconfigurations.

Modern backend frameworks often provide convenient ways to manage CSP headers without manually concatenating strings.

• Spring Security (Java): Has dedicated configurations for HTTP security headers, including CSP.
• Helmet (Node.js/Express): A middleware that helps secure Express apps by setting various HTTP headers, including CSP.
• Django (Python): Can set CSP headers via middleware or specific libraries.

These tools simplify implementation and help ensure best practices.

From a backend perspective, correctly configured CSP adds a powerful layer of defense against XSS. Even if an attacker manages to inject malicious content into your HTML, CSP can prevent the browser from executing it.

By controlling the Content-Security-Policy header, your backend dictates what content is safe, significantly reducing the impact of client-side vulnerabilities.

You've learned that Content Security Policy (CSP) is a critical security layer delivered by the backend via the Content-Security-Policy HTTP header.

• Backend sets CSP headers to control resource loading.
• Directives like script-src define allowed sources.
• Nonces are backend-generated tokens to safely allow specific inline scripts.
• Backend can collect violation reports via report-to.
• Frameworks simplify CSP implementation.

By actively managing CSP, backend developers significantly enhance their application's defense against client-side attacks like XSS.