Files
cinny/src/app/utils/linkSafety.test.ts
T
Lotus CIandClaude Opus 5.5 a8a7e639a9
CI / Build & Quality Checks (pull_request) Successful in 1m47s
CI / Trigger Desktop Build (pull_request) Skipped
CI / Docker image build & smoke test (pull_request) Skipped
CI / Secret scan (gitleaks) (pull_request) Successful in 7s
CI / Playwright smoke (e2e) (pull_request) Failing after 11m30s
fix(security): link warning no longer bypassed by a sentence or shared hosting (#335)
Two easy bypasses of the "this link goes somewhere else" check:

- Any whitespace in the link text disabled it, so "Log in at
  chat.lotusguild.org" → evil.example wasn't flagged. hostsFromText now
  scans each word for host-shaped tokens; a link is flagged if any named
  host is a different site. Bare file names (main.py, report.pdf) and
  version numbers are skipped inside sentences so ordinary links aren't
  flagged.
- Shared-hosting suffixes counted as one site (alice.github.io vs
  evil.github.io, *.pages.dev, netlify.app, vercel.app, …).
  registrableDomain now treats a curated list of them as public suffixes
  (longest match, incl. s3.amazonaws.com). Curated rather than the full
  PSL to stay inside the bundle budget.

Verified with unit tests and headless on real messages: the four bypass
cases are flagged; the honest sentence, file/version text and "click
here" are not.

Fixes #335

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017PCb59uY3UKqDqt7B6hzd5
2026-10-16 18:56:01 -04:00

76 lines
3.5 KiB
TypeScript
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
import { test } from 'node:test';
import assert from 'node:assert/strict';
import { analyzeLink, hostFromText, registrableDomain } from './linkSafety';
test('visible text that names another site is a mismatch', () => {
const r = analyzeLink('https://matrix.lotusguild.org/login', 'https://evil.example/login');
assert.equal(r?.mismatch, true);
assert.equal(r?.shownHost, 'matrix.lotusguild.org');
assert.equal(r?.realHost, 'evil.example');
});
test('honest links are not flagged: same registrable domain, plain words, non-http', () => {
assert.equal(
analyzeLink('youtube.com/watch?v=1', 'https://www.youtube.com/watch?v=1')?.mismatch,
false,
);
assert.equal(analyzeLink('bbc.co.uk', 'https://news.bbc.co.uk/x')?.mismatch, false);
assert.equal(analyzeLink('click here', 'https://evil.example')?.mismatch, false);
assert.equal(analyzeLink('evil.example', 'mailto:someone@evil.example'), null);
assert.equal(analyzeLink('elsewhere.org', 'https://matrix.to/#/#room:x'), null);
assert.equal(
analyzeLink('lotusguild.org', 'https://chat.lotusguild.org/home/!r:x')?.mismatch,
false,
);
});
test('registrable domain handles two-label suffixes', () => {
assert.equal(registrableDomain('news.bbc.co.uk'), 'bbc.co.uk');
assert.equal(registrableDomain('www.example.com'), 'example.com');
assert.equal(registrableDomain('example.com'), 'example.com');
});
test('hostFromText only accepts URL/host-shaped text; IDN becomes punycode', () => {
assert.equal(hostFromText('Visit https://a.example/path'), null);
assert.equal(hostFromText('a.example/path'), 'a.example');
assert.equal(hostFromText('user@a.example:8448/x'), 'a.example');
assert.equal(hostFromText('pаypal.com'), 'xn--pypal-4ve.com'); // Cyrillic а
assert.equal(analyzeLink('paypal.com', 'https://xn--pypal-4ve.com/')?.punycode, true);
assert.equal(analyzeLink('paypal.com', 'https://xn--pypal-4ve.com/')?.mismatch, true);
});
test('a trusted name inside a sentence still exposes a different destination (#335)', () => {
const a = analyzeLink('Log in at chat.lotusguild.org', 'https://evil.example/');
assert.equal(a?.mismatch, true);
assert.equal(a?.shownHost, 'chat.lotusguild.org');
assert.equal(
analyzeLink('chat.lotusguild.org (official)', 'https://evil.example/')?.mismatch,
true,
);
assert.equal(
analyzeLink('Open https://paypal.com/login now', 'https://evil.example/')?.mismatch,
true,
);
// honest sentences are not flagged
assert.equal(
analyzeLink('Log in at chat.lotusguild.org', 'https://chat.lotusguild.org/login')?.mismatch,
false,
);
assert.equal(analyzeLink('read the docs', 'https://evil.example/')?.mismatch, false);
assert.equal(analyzeLink('see main.py and report.pdf', 'https://github.com/x')?.mismatch, false);
assert.equal(analyzeLink('release v1.2.3 notes', 'https://github.com/x')?.mismatch, false);
});
test('shared hosting suffixes are separate sites (#335)', () => {
assert.equal(registrableDomain('alice.github.io'), 'alice.github.io');
assert.equal(registrableDomain('www.alice.github.io'), 'alice.github.io');
assert.equal(registrableDomain('bucket.s3.amazonaws.com'), 'bucket.s3.amazonaws.com');
assert.equal(analyzeLink('alice.github.io', 'https://evil.github.io/')?.mismatch, true);
assert.equal(analyzeLink('lotusguild.pages.dev', 'https://evil.pages.dev/')?.mismatch, true);
assert.equal(
analyzeLink('alice.github.io/blog', 'https://alice.github.io/blog/1')?.mismatch,
false,
);
assert.equal(registrableDomain('news.bbc.co.uk'), 'bbc.co.uk');
});