การรักษาความปลอดภัยแอประยะไกลของ Module Federation

Module Federation fetches and executes remote JavaScript at run time. That power is also a risk: if an attacker controls a remote URL, they can run code inside your app.

Key threats to the federation layer include:

• A compromised remote host serving malicious code
• Man-in-the-middle modification of remoteEntry.js
• Loading a remote from an unexpected origin

Loading any remote over plain HTTP allows in-transit tampering. Every remoteEntry.js and chunk must be served over HTTPS, with HSTS enforced.

Do not load remotes from arbitrary URLs. Restrict allowed origins with a Content Security Policy so only trusted hosts can supply scripts.

SRI lets the browser verify a fetched script matches a known hash, rejecting it if it was altered. Pairing SRI with federation guards against tampered remotes.

If you load remote URLs from a manifest, that manifest is a high-value target. Serve it from a trusted origin and validate its contents before using any URL.

Never build a remote URL from user input or untrusted config. An attacker who influences the URL can point your app at malicious code.

Because remotes share the same page context, a malicious remote can read the DOM and globals. For untrusted third-party MFEs, consider iframe or sandbox isolation.

A compromised remote can read shared stores and globals. Never place raw auth tokens on window or in shared state where any remote could harvest them.

Generate and pin SRI hashes during the build, and check that deployed remoteEntry files match expected hashes, so a tampered artifact fails verification before users hit it.

No single control is enough. Combine HTTPS, CSP allowlists, SRI, manifest validation, and isolation so that bypassing one layer still leaves others protecting the app.

Test your federation-security knowledge.

You learned to secure federation remotes:

• Remotes execute live code, so the loader is an attack surface
• Always use HTTPS and a CSP script-src allowlist
• Verify integrity with SRI and hash checks in CI
• Never load remotes from untrusted URLs
• Isolate untrusted MFEs and protect tokens

Defense in depth keeps federated code trustworthy.