Files
openclaw/src/link-understanding/detect.test.ts
T
ly-wang19 d4a01e48bc fix(link-understanding): strip markdown links whose label contains brackets (#96476)
Merged via squash.

Prepared head SHA: 2d69ed259f
Co-authored-by: ly-wang19 <94427531+ly-wang19@users.noreply.github.com>
Co-authored-by: vincentkoc <25068+vincentkoc@users.noreply.github.com>
Reviewed-by: @vincentkoc
2026-06-27 05:43:11 +08:00

81 lines
3.6 KiB
TypeScript

// Link detection tests cover ordering, dedupe, markdown suppression, and SSRF hostname filtering.
import { describe, expect, it } from "vitest";
import { extractLinksFromMessage } from "./detect.js";
describe("extractLinksFromMessage", () => {
it("extracts bare http/https URLs in order", () => {
const links = extractLinksFromMessage("see https://a.example and http://b.test");
expect(links).toEqual(["https://a.example", "http://b.test"]);
});
it("dedupes links and enforces maxLinks", () => {
const links = extractLinksFromMessage("https://a.example https://a.example https://b.test", {
maxLinks: 1,
});
expect(links).toEqual(["https://a.example"]);
});
it("ignores markdown links", () => {
const links = extractLinksFromMessage("[doc](https://docs.example) https://bare.example");
expect(links).toEqual(["https://bare.example"]);
});
it("ignores markdown links whose label contains brackets", () => {
// The closing "]" inside the label must not break markdown stripping, otherwise
// the citation URL leaks out as a bare link (with a stray trailing ")").
const links = extractLinksFromMessage(
"Check [my notes [v2]](https://internal.example/doc) for details",
);
expect(links).toStrictEqual([]);
});
it("blocks 127.0.0.1", () => {
const links = extractLinksFromMessage("http://127.0.0.1/test https://ok.test");
expect(links).toEqual(["https://ok.test"]);
});
it("blocks localhost and common loopback addresses", () => {
expect(extractLinksFromMessage("http://localhost/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://localhost.localdomain/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://foo.localhost/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://service.local/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://service.internal/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://0.0.0.0/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://[::1]/secret")).toStrictEqual([]);
});
it("blocks private network ranges", () => {
expect(extractLinksFromMessage("http://10.0.0.1/internal")).toStrictEqual([]);
expect(extractLinksFromMessage("http://172.16.0.1/internal")).toStrictEqual([]);
expect(extractLinksFromMessage("http://192.168.1.1/internal")).toStrictEqual([]);
});
it("blocks link-local and cloud metadata addresses", () => {
expect(extractLinksFromMessage("http://169.254.169.254/latest/meta-data/")).toStrictEqual([]);
expect(extractLinksFromMessage("http://169.254.1.1/test")).toStrictEqual([]);
expect(extractLinksFromMessage("http://metadata.google.internal/computeMetadata/v1/")).toEqual(
[],
);
});
it("blocks CGNAT range used by Tailscale", () => {
expect(extractLinksFromMessage("http://100.100.50.1/test")).toStrictEqual([]);
});
it("blocks private and mapped IPv6 addresses", () => {
expect(extractLinksFromMessage("http://[::ffff:127.0.0.1]/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://[2001:db8:1234::5efe:127.0.0.1]/secret")).toStrictEqual(
[],
);
expect(extractLinksFromMessage("http://[fe80::1]/secret")).toStrictEqual([]);
expect(extractLinksFromMessage("http://[fc00::1]/secret")).toStrictEqual([]);
});
it("allows legitimate public URLs", () => {
expect(extractLinksFromMessage("https://example.com/page")).toEqual([
"https://example.com/page",
]);
expect(extractLinksFromMessage("https://8.8.8.8/dns")).toEqual(["https://8.8.8.8/dns"]);
});
});