Supercharge your AI agents with data from the web and beyond. Building the library for superintelligence. 🔥
TypeScript aiai-agentsai-crawlerai-scrapingai-searchcrawlerdata-extractionhtml-to-markdownllmmarkdownscraperscrapingweb-crawlerweb-dataweb-data-extractionweb-scraperweb-scrapingweb-searchwebscraping
import { ALLOW_TEST_SUITE_WEBSITE, itIf, TEST_SUITE_WEBSITE } from "../lib";
import request, { idmux, Identity, scrapeTimeout, TEST_API_URL } from "./lib";
describe("Hosted bash", () => {
let identity: Identity;
beforeAll(async () => {
identity = await idmux({ name: "bash", concurrency: 10, credits: 1000 });
}, 10000);
const run = (command: string) =>
request(TEST_API_URL)
.post("/v2/bash")
.set("Authorization", `Bearer ${identity.apiKey}`)
.type("form")
.send(command);
it("serves usage", async () => {
const res = await request(TEST_API_URL).get("/v2/bash");
expect(res.status).toBe(200);
expect(res.text).toContain("/v2/bash --data-binary 'firecrawl search");
});
it("runs a command in a fresh shell", async () => {
const res = await run("echo hello | tr a-z A-Z; cat /README.md | head -1");
expect(res.status).toBe(200);
expect(res.headers["x-exit-code"]).toBe("0");
expect(res.text).toBe("HELLO\n# Firecrawl bash\n");
});
itIf(ALLOW_TEST_SUITE_WEBSITE)(
"scrapes through the API's own scrape route",
async () => {
const res = await run(
`firecrawl scrape ${TEST_SUITE_WEBSITE} | grep -c Firecrawl`,
);
expect(res.headers["x-exit-code"]).toBe("0");
expect(Number(res.text.trim())).toBeGreaterThan(0);
},
scrapeTimeout,
);
it("exits 2 for a command outside the hosted subset", async () => {
const res = await run("firecrawl crawl https://example.com");
expect(res.headers["x-exit-code"]).toBe("2");
expect(res.text).toContain("npx -y firecrawl-cli@latest init");
});
it("stops a runaway command", async () => {
const res = await run("while true; do :; done");
expect(res.headers["x-exit-code"]).toBe("126");
});
});