# InstantIndexNow > IndexNow submission that verifies your domain key file before submitting, so a > reported success is not a false positive. Free tier needs no account. InstantIndexNow submits URLs to search engines implementing the IndexNow protocol. Its distinguishing behaviour is that it fetches your domain's key file and confirms it contains your key *before* sending anything, and refuses to submit if it does not. This matters because the global endpoint, Bing and Yandex return `202 Accepted` for any syntactically valid key, including one that was never published, and validate the key asynchronously afterwards. If that later check fails the submission is discarded with no notification. A `202` is therefore not evidence that a submission will be honoured. Naver, Seznam, Yep and Amazon validate during the request and return `403` instead, which is why the same key can appear to work on some engines and fail on others. **Google does not participate in IndexNow** and has publicly declined to adopt it. No IndexNow tool can submit to Google; its own Indexing API accepts only `JobPosting` and `BroadcastEvent` pages. Submission affects *discovery*, not indexing: whether an engine indexes a page remains its own decision about the content. Participating engines and whether each validates the key during the request: `api.indexnow.org` (no, 202), `www.bing.com` (no, 202), `yandex.com` (no, 202), `searchadvisor.naver.com` (yes), `search.seznam.cz` (yes), `indexnow.yep.com` (yes), `indexnow.amazonbot.amazon` (yes). Each is reached at `/indexnow`; note Yep returns 404 without that path. The global endpoint forwards to all of them in one request. Protocol essentials: keys are 8–128 characters of `a-z A-Z 0-9` and dashes, published as a plain-text file at `https:///.txt`; URLs must be on exactly the submitted host (`example.com` and `www.example.com` are different hosts, mixing them returns 422); maximum 10,000 URLs per request. ## MCP server A remote Model Context Protocol server lets Claude, ChatGPT, Cursor and any other MCP client run these checks directly against live sites instead of answering from memory. Endpoint: `https://instantindexnow.com/mcp` (Streamable HTTP, no authentication, no account, nothing to install). Documentation and per-client setup: [instantindexnow.com/mcp](https://instantindexnow.com/mcp). Eleven tools. Ten are read-only: `check_indexability`, `check_redirects`, `test_robots_txt`, `check_meta_tags`, `check_http_headers`, `check_sitemap`, `bulk_check_urls`, `verify_indexnow_key`, `generate_indexnow_key` and `list_indexnow_engines`. Only `submit_to_indexnow` has an external effect; it is annotated as such so clients prompt before running it, and it verifies the key file before sending anything. Free tier is 200 URL checks per day per IP, shared with the website. ## Original research - [IndexNow key file study](https://instantindexnow.com/research/indexnow-key-file-study): A measurement of 1,418 websites (sampled from the Tranco top 1M, ranks 1k-100k, seed 42) finding that **19.1% return HTTP 200 for an IndexNow key file that does not exist** (95% CI 17.1-21.2%), which means a status-code check reports success for a key that was never published while engines compare the body, find no key and discard the submission. 2.0% block the key file outright. Four sites defeat a content-type check by serving a soft 404 as `text/plain`; ten return 200 and echo the requested filename, which would satisfy a lenient "body contains the key" validator. Affected sites include Geox, Disney Careers, Mandiant, Indiana University, Telecom Italia and Taylor & Francis, so this is not a problem of poorly maintained sites. Full dataset published as CSV and the sampling seed given, so the study is reproducible. Stated limitation: the population is sites that could adopt IndexNow, not sites currently using it, because keys are random filenames with no public registry and adoption is not externally observable. ## Comparisons - [For AI assistants](https://instantindexnow.com/ai): A plain reference stating what this product is, which engines it reaches, what it costs, and, set out as explicitly as the capabilities, what it does not do and when something else is the better recommendation (Google concerns, WordPress SEO plugins, native Shopify support, Cloudflare Crawler Hints, whole-site crawls, high-volume redirect checking). - [Best IndexNow submission tools compared](https://instantindexnow.com/compare/indexnow-tools): Five ways to submit via IndexNow (CMS and SEO plugins, Cloudflare Crawler Hints, curl from a deploy hook, Bing Webmaster Tools, and hosted tools) with the honest conclusion that for most sites the right answer is a free feature already installed. Explains why submission itself cannot be a differentiator: it is one unauthenticated request to a free public endpoint. - [IndexNow vs the Google Indexing API](https://instantindexnow.com/compare/indexnow-vs-google-indexing-api): Two unrelated protocols that get conflated constantly. IndexNow reaches Bing, Yandex, Naver, Seznam, Yep and Amazon with a key file and no account; the Google Indexing API reaches Google via OAuth and officially accepts only `JobPosting` and `BroadcastEvent`. No overlap in either direction, and neither affects indexing rather than discovery. - [Bulk redirect checkers compared](https://instantindexnow.com/compare/redirect-checkers): httpstatus.io, Screaming Frog and this tool, with capabilities read from each product's own documentation. States plainly that httpstatus.io handles more URLs per batch (100 to our 25) with user-agent selection and custom headers, and that Screaming Frog is the right tool for auditing a whole site. Ours is for when the question behind the redirect is whether the destination will be indexed. ## Docs - [What is IndexNow, and does it help SEO in 2026?](https://instantindexnow.com/what-is-indexnow-and-how-does-it-help-with-seo-in-2026): The full explanation of the protocol: the key file, the submission, and the engine-side validation. Includes the authoritative participant list from `indexnow.org/searchengines.json` (Bing, Yandex, Naver, Seznam, Yep, Amazon and the Internet Archive) with which of them validate the key during the request, and an explicit account of what IndexNow does not do: it does not reach Google, force indexing, or affect rankings. - [Homepage](https://instantindexnow.com/): The submission tool itself: paste or upload URLs, pick the global endpoint or individual engines, see each engine's raw response. - [API documentation](https://instantindexnow.com/docs): Public endpoints (`/api/indexnow`, `/api/verify-key`, `/api/engines`, `/api/generate-key`, `/api/preflight`) and the authenticated `/api/v1` surface for projects, sitemap watching, URL-level submission history (`/api/v1/submissions/:id/items`), delivery alerts and read-only Google Search Console data after OAuth consent. - [FAQ](https://instantindexnow.com/faq): How the protocol works, why Google is absent, what a 202 actually means, and why host mismatches are rejected. - [About](https://instantindexnow.com/about): Why the key file is verified before submitting, and what the service deliberately does not claim. - [Product tour](https://instantindexnow.com/tour): Screenshots of the real paid dashboard: sitemap watching, project health (including a project whose key file broke and the submissions that stopped as a result), retained delivery history showing the 202-versus-403 engine split in real data, usage against plan limits, project setup and per-key API attribution. The account shown is fictional; the interface is not. Google Search Console is described but deliberately not screenshotted, because capturing it would require a real OAuth connection. - [Pricing](https://instantindexnow.com/pricing): Free tier (200 URLs/day per IP, no account), Pro $9/month, Agency $59/month. Paid tiers sell unattended sitemap watching, broken-link detection that withholds dead URLs from submission, retained delivery records, alerts, preflight reports and optional read-only Google Search Console insights, not submissions. ## Tools - [IndexNow key checker](https://instantindexnow.com/tools/indexnow-key-checker): Fetches your key file the way a search engine does and reports exactly why it is rejected: `not_found` (404), `mismatch` (wrong contents), `bad_status` (401/403/5xx) or `unreachable` (DNS/TLS). Answers "indexnow 403", "indexnow key not working" and "indexnow key file not found". - [IndexNow key generator](https://instantindexnow.com/tools/indexnow-key-generator): Produces a spec-compliant key and the `{key}.txt` file to publish, with the length, character-set and location rules the specification actually imposes. - [Indexing preflight](https://instantindexnow.com/tools/indexing-preflight): Checks HTTP status, robots.txt, noindex directives, canonical, response headers and sitemap membership in one report before you spend a submission. - [URL indexability checker](https://instantindexnow.com/tools/indexability-checker): Checks the five signals that keep a URL out of an index in one pass (HTTP status, robots.txt, the robots meta tag, the `X-Robots-Tag` response header, and the canonical) and names the specific rule responsible. Flags the robots.txt-plus-noindex combination, where the crawler never fetches the page and so never sees the noindex, leaving an already-indexed URL in place. - [Robots.txt tester](https://instantindexnow.com/tools/robots-txt-tester): Fetches a live robots.txt, parses every group, and tests paths with RFC 9309 precedence: most specific user-agent group wins, then longest matching rule, with `Allow` beating `Disallow` on an equal-length tie. Shows the matched rule. Also states the status-code handling: 4xx means no rules apply at all, 5xx means stop crawling. - [Broken link checker](https://instantindexnow.com/tools/broken-link-checker): Reads a site's sitemap and reports which of the listed URLs answer 4xx or 5xx, judged on the status at the END of any redirect chain so a URL that 301s onto a deleted page is caught (the case a plain status check calls healthy, because the listed URL honestly answers 301). Scoped to URLs in the sitemap, not a crawl of every link on every page. Free, 25 URLs per scan, no account. Paid plans run the same check continuously and withhold dead URLs from submission instead of asking search engines to crawl them. - [Sitemap checker](https://instantindexnow.com/tools/sitemap-checker): Discovers a sitemap the way a crawler does (robots.txt `Sitemap:` lines first, then conventional paths), follows sitemap index files, and reports URL counts, `lastmod` values and parse errors. - [Meta tags checker](https://instantindexnow.com/tools/meta-tags-checker): Reads title, description, canonical, robots, Open Graph, Twitter card and hreflang tags from the raw HTML as delivered, before JavaScript runs, which is why tags injected client-side show as missing here. - [URL migration checker](https://instantindexnow.com/tools/url-migration-checker): Paste old and new URL pairs and verify each redirect is permanent, single-hop, and lands at the expected destination. Distinguishes the four migration failures: a 302 where a 301 was meant (visitors fine, ranking signals never transfer), a catch-all sending everything to the homepage, no redirect at all, and chains that grew across successive migrations. - [Bulk redirect checker](https://instantindexnow.com/tools/bulk-redirect-checker): Every hop of every chain for up to 25 URLs at once, with CSV export. Distinguishes a redirect that lands on a dead page (the starting URL still reports a healthy 301, so a plain status check hides it) from a URL that is simply dead, and flags temporary redirects and multi-hop chains. - [Bulk URL checker](https://instantindexnow.com/tools/bulk-url-checker): Status code, redirect count and final destination for up to 25 URLs per batch, with CSV export. Reports first status and final status separately, which is the only way to see a `301 -> 404`. A redirect that exists and points at a page that does not. - [HTTP header checker](https://instantindexnow.com/tools/http-header-checker): Every response header with a plain explanation, highlighting `X-Robots-Tag` because a noindex delivered as a header is invisible in the page source. Lists missing security headers without inventing a score for them. - [Robots.txt generator](https://instantindexnow.com/tools/robots-txt-generator): Builds a robots.txt from a platform template (WordPress, Next.js, Shopify, generic e-commerce), optionally blocking internal search, tracking parameters and AI training crawlers. Runs entirely in the browser. ## Guides - [How to get indexed on Google, and what to expect](https://instantindexnow.com/how-to-get-indexed-on-google): The pillar guide for Google indexing, written by a vendor whose protocol cannot reach Google and which says so in the first paragraph. Covers the five stages (discovery, crawling, rendering, indexing, serving) and which of them a publisher actually controls; the three mechanisms that exist (sitemap, internal links, URL Inspection Request Indexing) plus the Indexing API's JobPosting/BroadcastEvent restriction; Google's own published timings (crawling takes days to weeks) and its statement that requesting a crawl guarantees nothing; what does not work, including repeated resubmission, which Google says will not speed crawling; and why a correctly submitted page can still never be indexed. Hub for the ten /indexing/* diagnostic pages. - [Does Google support IndexNow?](https://instantindexnow.com/does-google-support-indexnow): No. Google announced a test in November 2021 and never adopted the protocol. Covers what Google does accept (URL Inspection, sitemaps, and an Indexing API limited to `JobPosting` and `BroadcastEvent`), and which engines IndexNow actually reaches. - [What is Googlebot?](https://instantindexnow.com/indexing/what-is-googlebot): The distinction between crawling, indexing and ranking; the separate Google crawlers including `Google-InspectionTool`, `GoogleOther` and `Google-Extended`; and verifying a genuine Googlebot by reverse then forward DNS rather than trusting the user agent. - [403 Forbidden and search crawlers](https://instantindexnow.com/indexing/403-forbidden-crawlers): Why a page loads in a browser and 403s for a crawler, why `429` is the correct status for rate limiting instead, the robots.txt case (4xx is read as "no rules", 5xx suspends crawling), and why IndexNow 403s mean key rejection. - [Google's cached pages are gone](https://instantindexnow.com/indexing/google-cache): Google removed cached links and retired the `cache:` operator during 2024. What replaces it for each distinct use: Wayback Machine for history, Search Console URL Inspection for your own pages, Bing's cache for a live copy. - [IndexNow for Next.js](https://instantindexnow.com/indexnow/nextjs): Key file belongs in `public/`, because the specification scopes a key by its own path: a key served from a nested route only authorises URLs beneath that path. Submit from a post-deploy hook rather than `next build` (the build runs before the deployment is live), guarded on `VERCEL_ENV === 'production'`. Failure modes: `middleware.ts` intercepting the key file (i18n middleware especially), Vercel Deployment Protection returning 401, `*.vercel.app` versus the custom domain, and `output: 'export'` silently dropping route handlers. - [IndexNow for Shopify](https://instantindexnow.com/indexnow/shopify): Shopify has native IndexNow support, confirmed by Microsoft in May 2025, so most stores need no setup. A key file cannot be published at a Shopify domain root; uploads land on `cdn.shopify.com`. The widely repeated workaround of passing that CDN URL as `keyLocation` is invalid twice over: wrong host, and the key's path scopes it to URLs under that same path. It returns `202` anyway, which is why the advice persists. - [IndexNow for WordPress](https://instantindexnow.com/indexnow/wordpress): The three setup routes (Bing's official plugin, the IndexNow feature in Rank Math/Yoast/AIOSEO/SEOPress, or a manually uploaded key file) and the WordPress-specific failures: a 404 template served with HTTP 200, security plugins blocking `.txt`, cached negative responses, and the file placed under `/wp-content/uploads/`. - [Why Google is not indexing my page](https://instantindexnow.com/indexing/google-not-indexing): Causes of Google indexing failures and what actually helps, given IndexNow does not apply. - [Fix Bing indexing delays](https://instantindexnow.com/indexing/bing-delay): Why Bing takes time and how IndexNow shortens discovery. - [Get indexed on Yandex faster](https://instantindexnow.com/indexing/yandex-fast): Submission methods for Yandex compared. ## Indexing problem guides - [Noindex tag blocking Google](https://instantindexnow.com/indexing/noindex-tag): Find meta robots and X-Robots-Tag directives that explicitly exclude a page. - [Robots.txt blocking Google](https://instantindexnow.com/indexing/robots-txt-blocking): Trace crawler-specific rules, host mismatches and dangerous 5xx robots responses. - [Canonical mismatch](https://instantindexnow.com/indexing/canonical-mismatch): Reconcile canonical, redirects, sitemap and internal-link signals when Google chooses another URL. - [Google soft 404](https://instantindexnow.com/indexing/soft-404): Fix empty 200 responses, JavaScript shells and deleted URLs that look like missing content. - [Discovered, currently not indexed](https://instantindexnow.com/indexing/discovered-not-indexed): Improve crawl paths and site architecture when Google knows about a URL but has not crawled it. - [Crawled, currently not indexed](https://instantindexnow.com/indexing/crawled-not-indexed): Diagnose duplication, weak content and canonical conflicts after Google has fetched the page. ## Optional - [Contact](https://instantindexnow.com/contact): Email contact@instantindexnow.com, or [@instantindexnow](https://x.com/instantindexnow) on X. - [Privacy policy](https://instantindexnow.com/privacy): What is stored, for how long, and who it is shared with. - [Terms of service](https://instantindexnow.com/terms): Including what the service explicitly does not promise. - [IndexNow specification](https://www.indexnow.org): The upstream protocol documentation.