{"id":24979,"date":"2026-08-06T10:17:40","date_gmt":"2026-08-06T10:17:40","guid":{"rendered":"https:\/\/capskip.com\/?p=24979"},"modified":"2026-08-06T10:17:40","modified_gmt":"2026-08-06T10:17:40","slug":"geetest-v3-nodejs","status":"publish","type":"post","link":"https:\/\/capskip.com\/zh\/geetest-v3-nodejs\/","title":{"rendered":"\u5982\u4f55\u7528 Node.js \u8bc6\u522b\u6781\u9a8c v3 \u5e76\u56de\u4f20"},"content":{"rendered":"

Almost every CAPTCHA integration assumes one token goes in and one comes out. GeeTest does not work that way. A solve returns three<\/strong> values that must be submitted together, and the challenge you started from expires about a minute after you fetched it. Neither failure produces a useful error message, which is why GeeTest has a reputation for being fiddly.<\/p>\n

Two inputs, one of them perishable<\/h2>\n\n\n\n\n\n\n
Value<\/th>\nLifetime<\/th>\n<\/tr>\n<\/thead>\n
gt<\/code><\/td>\nStatic per site. Cache it freely<\/td>\n<\/tr>\n
challenge<\/code><\/td>\nSingle use, dead in roughly 60 seconds<\/strong><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Both come from the site’s own GeeTest init endpoint. Treating the challenge as reusable is the reason integrations pass in testing and collapse in production, where queueing delay pushes the solve past the expiry.<\/p>\n

Setup<\/h2>\n
npm install capskip<\/pre>\n
const { CapSkip } = require('capskip');\n\nconst solver = new CapSkip({\n  host: '127.0.0.1',\n  port: 8080,\n  recaptchaTimeout: 300,   \/\/ GeeTest uses this, not defaultTimeout\n});<\/pre>\n
That distinction matters. defaultTimeout<\/code> only governs image CAPTCHAs; everything interactive uses recaptchaTimeout<\/code>.<\/p>\n
Solving<\/h2>\n
const result = await solver.geetest(\n  '81388ea1fc187e0c335c0a8907ff2625',   \/\/ gt, static per site\n  '7cf6a8b1a2c34d5e6f7089abcdef0123',   \/\/ challenge, fetched seconds ago\n  'https:\/\/example.com\/login',\n);\n\nconsole.log(result.challenge, result.validate, result.seccode);<\/pre>\n
Those three properties are the answer. result.code<\/code> is populated too, but for GeeTest it holds the raw JSON string rather than anything submittable, so reaching for it out of habit produces a baffling failure.<\/p>\n
Use the challenge<\/code> that comes back, not the one you passed in. They are not always identical.<\/p>\n
The whole flow<\/h2>\n
const { CapSkip } = require('capskip');\n\nconst solver = new CapSkip();\nconst LOGIN = 'https:\/\/example.com\/login';\n\n\/\/ 1. Fresh pair, cache-busted. Cached init responses return spent challenges.\nconst init = await (await fetch(\n  `https:\/\/example.com\/geetest\/init?t=${Date.now()}`,\n)).json();\n\n\/\/ 2. Solve immediately. Nothing slow between here and the previous step.\nconst result = await solver.geetest(init.gt, init.challenge, LOGIN);\n\n\/\/ 3. Post all three together with the real form fields.\nconst response = await fetch(LOGIN, {\n  method: 'POST',\n  body: new URLSearchParams({\n    geetest_challenge: result.challenge,\n    geetest_validate: result.validate,\n    geetest_seccode: result.seccode,\n    username: '...',\n    password: '...',\n  }),\n});<\/pre>\n
The cache-busting timestamp is doing real work. GeeTest init endpoints are frequently cached by a CDN or proxy, and a cached response hands you a challenge somebody already used.<\/p>\n
Those field names are the usual GeeTest v3 convention, though a site can rename them. Check the real form before assuming.<\/p>\n
The batching trap<\/h2>\n
This is where Node’s ergonomics work against you. The obvious parallel version is wrong:<\/p>\n
\/\/ WRONG: every challenge is fetched up front, so the later solves\n\/\/ start against pairs that have already expired.\nconst pairs = await Promise.all(urls.map(fetchPair));\nconst results = await Promise.all(\n  pairs.map((p, i) => solver.geetest(p.gt, p.challenge, urls[i])));<\/pre>\n
Keep the fetch and the solve inside the same task so each pair is used within seconds of being issued:<\/p>\n
\/\/ RIGHT: each task fetches its own pair immediately before solving.\nconst results = await Promise.all(urls.map(async (url) => {\n  const pair = await fetchPair(url);\n  return solver.geetest(pair.gt, pair.challenge, url);\n}));<\/pre>\n
AsyncCapSkip<\/code> is exported as an alias of CapSkip<\/code> here, so there is no separate async client to reach for. Promise.all<\/code> is the whole story.<\/p>\n
Errors<\/h2>\n
const {\n  ValidationException, NetworkException, ApiException, TimeoutException,\n} = require('capskip');\n\ntry {\n  const result = await solver.geetest(gt, challenge, pageUrl);\n} catch (err) {\n  if (err instanceof ValidationException)   { \/* missing gt or challenge *\/ }\n  else if (err instanceof NetworkException) { \/* CapSkip not running *\/ }\n  else if (err instanceof ApiException)     { \/* usually an expired challenge *\/ }\n  else if (err instanceof TimeoutException) { \/* exceeded recaptchaTimeout *\/ }\n  else throw err;\n}<\/pre>\n
Most GeeTest failures arrive as ApiException<\/code> and mean the challenge expired. Retrying with the same pair never works. Fetch a new one.<\/p>\n
Frequently asked questions<\/h2>\n
\n\n
Why can I not just use result.code?<\/h3>\n<\/summary>\n
Because GeeTest’s answer is three values. code<\/code> holds the raw JSON for completeness while the SDK expands the useful parts into challenge<\/code>, validate<\/code> and seccode<\/code>. Submit those.<\/p>\n<\/details>\n
\n\n
Can I cache the gt value?<\/h3>\n<\/summary>\n
Yes. The gt<\/code> is a site identifier and rarely changes. The challenge<\/code> is the perishable half and must be fetched fresh every time.<\/p>\n<\/details>\n
\n\n
Does this handle GeeTest v4?<\/h3>\n<\/summary>\n
The geetest<\/code> method targets v3, the slide puzzle built on a gt<\/code> and challenge<\/code> pair. v4 changed the parameter model, so check the current API documentation<\/a> before assuming the same call works.<\/p>\n<\/details>\n
Summary<\/h2>\n
Fetch a cache-busted pair, solve immediately, then post challenge<\/code>, validate<\/code> and seccode<\/code> together. Use the returned challenge, cache only the gt<\/code>, and keep fetch and solve inside the same task when running in parallel.<\/p>\n
Other languages are on the GeeTest solver<\/a> page, the wider Node surface on the Node.js CAPTCHA solver<\/a> page, and there is a live puzzle on our GeeTest v3 demo<\/a>. CapSkip does captcha bypass<\/a> locally, so solve volume costs nothing.<\/p>\n","protected":false},"excerpt":{"rendered":"
\u6781\u9a8c\u8fd4\u56de\u4e09\u4e2a\u503c\u800c\u975e token\uff0c\u4e14\u6311\u6218\u7ea6\u4e00\u5206\u949f\u540e\u5931\u6548\u3002\u4e0b\u9762\u662f Node.js \u6d41\u7a0b\uff0c\u5305\u62ec\u4e3a\u4ec0\u4e48\u6279\u5904\u7406\u4f1a\u7834\u574f\u5b83\u3002<\/p>","protected":false},"author":1,"featured_media":24978,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"rank_math_title":"Solve GeeTest v3 in Node.js | CapSkip","rank_math_description":"GeeTest returns three values in Node.js rather than one token, and the challenge expires in about a minute. Here is the full flow and the batching trap.","rank_math_focus_keyword":"solve geetest in node.js","footnotes":""},"categories":[70],"tags":[],"class_list":["post-24979","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-captcha"],"_links":{"self":[{"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/posts\/24979","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/comments?post=24979"}],"version-history":[{"count":1,"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/posts\/24979\/revisions"}],"predecessor-version":[{"id":24995,"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/posts\/24979\/revisions\/24995"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/media\/24978"}],"wp:attachment":[{"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/media?parent=24979"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/categories?post=24979"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/capskip.com\/zh\/wp-json\/wp\/v2\/tags?post=24979"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}