[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"blog-post-en-ja4-fingerprint-decoded":3},{"success":4,"payload":5,"related":77,"errors":158},true,{"id":6,"slug":7,"title":8,"excerpt":9,"body_html":10,"meta_title":11,"meta_description":12,"canonical_url":13,"canonical_override":14,"og_title":15,"og_description":12,"og_image_url":16,"faq_json":17,"key_takeaways":33,"allow_index":4,"noindex_google_only":39,"allow_follow":4,"published_at":40,"updated_at":41,"content_updated_at":42,"featured_image_url":43,"featured_image_alt":8,"card_image_url":44,"reading_time_minutes":45,"author":46,"categories":50,"tags":57,"alternates":69,"locale_paths":75,"views":76},207,"ja4-fingerprint-decoded","JA4 Fingerprint Decoded: What t13d1516h2 and b0da82dd1658 Mean","b0da82dd1658 is the third section of a Chromium JA4: a resumed TLS 1.3 session with the old ALPS codepoint. Here is how to decode every part of a JA4 string, with real values you can reproduce.","\u003Cp>A JA4 fingerprint is a three-part string, \u003Ccode>a_b_c\u003C\u002Fcode>, computed from a TLS ClientHello. Part \u003Ccode>a\u003C\u002Fcode> is readable metadata (protocol, TLS version, SNI, cipher and extension counts, ALPN), part \u003Ccode>b\u003C\u002Fcode> is a 12-character SHA-256 of the sorted cipher list, and part \u003Ccode>c\u003C\u002Fcode> is a 12-character SHA-256 of the sorted extensions plus signature algorithms. If you searched for \u003Ccode>b0da82dd1658\u003C\u002Fcode>: it is a \u003Ccode>c\u003C\u002Fcode> section, and it belongs to Chromium. Specifically, it is what a Chromium-based client produces when it \u003Cem>resumes\u003C\u002Fem> a TLS 1.3 session while still using the older ALPS extension codepoint. The rest of this page shows how we know that, and how to decode any JA4 string yourself.\u003C\u002Fp>\n\n\u003Ch2>How a JA4 fingerprint is built: the a_b_c format\u003C\u002Fh2>\n\u003Cp>The \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4\u002Fblob\u002Fmain\u002Ftechnical_details\u002FJA4.md\" rel=\"noopener noreferrer\">official JA4 technical specification\u003C\u002Fa> from FoxIO defines the fingerprint as one human-readable prefix and two truncated hashes, joined by underscores. GREASE values are ignored everywhere, and all hashes are lowercase.\u003C\u002Fp>\n\n\u003Ctable>\n\u003Cthead>\u003Ctr>\u003Cth>Section\u003C\u002Fth>\u003Cth>Characters\u003C\u002Fth>\u003Cth>What it encodes\u003C\u002Fth>\u003Cth>Possible values\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\n\u003Ctbody>\n\u003Ctr>\u003Ctd>a\u003C\u002Ftd>\u003Ctd>1\u003C\u002Ftd>\u003Ctd>Transport\u003C\u002Ftd>\u003Ctd>\u003Ccode>t\u003C\u002Fcode> = TLS over TCP, \u003Ccode>q\u003C\u002Fcode> = QUIC, \u003Ccode>d\u003C\u002Fcode> = DTLS\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>a\u003C\u002Ftd>\u003Ctd>2\u003C\u002Ftd>\u003Ctd>Highest TLS version (from \u003Ccode>supported_versions\u003C\u002Fcode> if present, else the ClientHello version)\u003C\u002Ftd>\u003Ctd>\u003Ccode>13\u003C\u002Fcode>, \u003Ccode>12\u003C\u002Fcode>, \u003Ccode>11\u003C\u002Fcode>, \u003Ccode>10\u003C\u002Fcode>, \u003Ccode>s3\u003C\u002Fcode>, \u003Ccode>d2\u003C\u002Fcode>, \u003Ccode>00\u003C\u002Fcode> = unknown…\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>a\u003C\u002Ftd>\u003Ctd>1\u003C\u002Ftd>\u003Ctd>SNI present?\u003C\u002Ftd>\u003Ctd>\u003Ccode>d\u003C\u002Fcode> = domain (SNI sent), \u003Ccode>i\u003C\u002Fcode> = no SNI (usually connecting to a bare IP)\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>a\u003C\u002Ftd>\u003Ctd>2\u003C\u002Ftd>\u003Ctd>Number of cipher suites (GREASE excluded)\u003C\u002Ftd>\u003Ctd>\u003Ccode>00\u003C\u002Fcode>–\u003Ccode>99\u003C\u002Fcode>\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>a\u003C\u002Ftd>\u003Ctd>2\u003C\u002Ftd>\u003Ctd>Number of extensions (GREASE excluded, SNI and ALPN included)\u003C\u002Ftd>\u003Ctd>\u003Ccode>00\u003C\u002Fcode>–\u003Ccode>99\u003C\u002Fcode>\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>a\u003C\u002Ftd>\u003Ctd>2\u003C\u002Ftd>\u003Ctd>First and last character of the first ALPN value\u003C\u002Ftd>\u003Ctd>\u003Ccode>h2\u003C\u002Fcode>, \u003Ccode>h1\u003C\u002Fcode> (http\u002F1.1), \u003Ccode>h3\u003C\u002Fcode>, \u003Ccode>00\u003C\u002Fcode> = no ALPN\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>b\u003C\u002Ftd>\u003Ctd>12\u003C\u002Ftd>\u003Ctd>SHA-256 of cipher suites as 4-char hex, sorted, comma-joined\u003C\u002Ftd>\u003Ctd>\u003Ccode>000000000000\u003C\u002Fcode> if none\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>c\u003C\u002Ftd>\u003Ctd>12\u003C\u002Ftd>\u003Ctd>SHA-256 of extensions (sorted, SNI \u003Ccode>0000\u003C\u002Fcode> and ALPN \u003Ccode>0010\u003C\u002Fcode> removed) + \u003Ccode>_\u003C\u002Fcode> + signature algorithms in wire order\u003C\u002Ftd>\u003Ctd>\u003Ccode>000000000000\u003C\u002Fcode> if none\u003C\u002Ftd>\u003C\u002Ftr>\n\u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n\u003Cp>Two details trip people up. First, the extension \u003Cem>count\u003C\u002Fem> in part a includes SNI and ALPN, but the extension \u003Cem>hash\u003C\u002Fem> in part c excludes them, so the same client produces the same c whether it connects to a hostname or an IP. Second, signature algorithms are hashed in the order the client sent them, not sorted, so a client that reorders its signature preferences changes c even with identical extensions.\u003C\u002Fp>\n\n\u003Ch3>Decoding t13d1516h2_8daaf6152771_… piece by piece\u003C\u002Fh3>\n\u003Cul>\n\u003Cli>\u003Ccode>t\u003C\u002Fcode>: TLS over TCP, not QUIC.\u003C\u002Fli>\n\u003Cli>\u003Ccode>13\u003C\u002Fcode>: the client offers TLS 1.3 as its highest version.\u003C\u002Fli>\n\u003Cli>\u003Ccode>d\u003C\u002Fcode>: an SNI extension was present, so the client was connecting to a domain name.\u003C\u002Fli>\n\u003Cli>\u003Ccode>15\u003C\u002Fcode>: 15 cipher suites after removing GREASE.\u003C\u002Fli>\n\u003Cli>\u003Ccode>16\u003C\u002Fcode>: 16 extensions after removing GREASE, counting SNI and ALPN.\u003C\u002Fli>\n\u003Cli>\u003Ccode>h2\u003C\u002Fcode>: the first ALPN value is \u003Ccode>h2\u003C\u002Fcode>, meaning the client prefers HTTP\u002F2.\u003C\u002Fli>\n\u003Cli>\u003Ccode>8daaf6152771\u003C\u002Fcode>: the hash of this sorted cipher string, which you can reproduce with any SHA-256 tool:\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cpre>\u003Ccode class=\"language-bash\">printf '%s' \"002f,0035,009c,009d,1301,1302,1303,c013,c014,c02b,c02c,c02f,c030,cca8,cca9\" \\\n  | sha256sum | cut -c1-12\n# 8daaf6152771\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>That list is the three TLS 1.3 suites plus Chrome's twelve legacy TLS 1.2 suites. Every Chromium build we checked, from the FoxIO examples through a headless Chrome 148, still produces \u003Ccode>8daaf6152771\u003C\u002Fcode>. So the b section tells you \"Chromium-family TLS stack\" and almost nothing about the version. Version differences show up in c.\u003C\u002Fp>\n\u003Cp>One small caveat if you are reading the spec closely: its algorithm summary shows \u003Ccode>t13d1516h2_8daaf6152771_b186095e22b6\u003C\u002Fcode>, while its worked example hashes to \u003Ccode>e5627efa2ab1\u003C\u002Fcode>. We recomputed the worked example and got \u003Ccode>e5627efa2ab1\u003C\u002Fcode>, so treat the first string as illustrative.\u003C\u002Fp>\n\n\u003Ch2>What b0da82dd1658 means\u003C\u002Fh2>\n\u003Cp>\u003Ccode>b0da82dd1658\u003C\u002Fcode> is the c section of a Chromium JA4. It is the hash of this exact string:\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-text\">0005,000a,000b,000d,0012,0017,001b,0023,0029,002b,002d,0033,4469,fe0d,ff01_0403,0804,0401,0503,0805,0501,0806,0601\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>You can check it yourself:\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-bash\">printf '%s' \"0005,000a,000b,000d,0012,0017,001b,0023,0029,002b,002d,0033,4469,fe0d,ff01_0403,0804,0401,0503,0805,0501,0806,0601\" \\\n  | sha256sum | cut -c1-12\n# b0da82dd1658\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>Three extensions in that list identify the client:\u003C\u002Fp>\n\u003Cul>\n\u003Cli>\u003Ccode>0029\u003C\u002Fcode> (pre_shared_key). This is TLS 1.3 session resumption. It only appears when the client already has a ticket from an earlier connection to the same server. \u003Ca href=\"https:\u002F\u002Fwww.rfc-editor.org\u002Frfc\u002Frfc8446#section-4.2.11\" rel=\"noopener noreferrer\">RFC 8446 §4.2.11\u003C\u002Fa> defines it. Without it, the same browser hashes to \u003Ccode>02713d6af862\u003C\u002Fcode>.\u003C\u002Fli>\n\u003Cli>\u003Ccode>4469\u003C\u002Fcode> (ALPS, original codepoint). Application-Layer Protocol Settings is a Chrome extension. BoringSSL \u003Ca href=\"https:\u002F\u002Fboringssl.googlesource.com\u002Fboringssl\u002F+\u002F558960d1e1fabfdb508e4da7940185c58985c961%5E%21\u002F\" rel=\"noopener noreferrer\">added a new codepoint for it in September 2023\u003C\u002Fa> (\u003Ccode>0x44cd\u003C\u002Fcode>), and current Chrome sends the new one. So \u003Ccode>4469\u003C\u002Fcode> points to an older Chromium build or a tool that copies one.\u003C\u002Fli>\n\u003Cli>\u003Ccode>fe0d\u003C\u002Fcode> (Encrypted Client Hello). Chrome sends a GREASE ECH extension on ordinary connections.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cp>FoxIO's own \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4\u002Fblob\u002Fmain\u002Fja4plus-mapping.csv\" rel=\"noopener noreferrer\">ja4plus-mapping.csv\u003C\u002Fa> labels \u003Ccode>t13d1517h2_8daaf6152771_b0da82dd1658\u003C\u002Fcode> and \u003Ccode>t13i1516h2_8daaf6152771_b0da82dd1658\u003C\u002Fcode> as \"Chromium Browser\". The prefix arithmetic matches: 15 hashed extensions plus SNI plus ALPN gives \u003Ccode>17\u003C\u002Fcode> with SNI (\u003Ccode>d\u003C\u002Fcode>), and 16 without it (\u003Ccode>i\u003C\u002Fcode>). That also exposes an inconsistency you'll see repeated online. A string like \u003Ccode>t13d1516h2_8daaf6152771_b0da82dd1658\u003C\u002Fcode> cannot come from a single real ClientHello: with SNI and ALPN both present, those 15 hashed extensions would have to produce a count of 17.\u003C\u002Fp>\n\u003Cp>We also reproduced the value directly. On 2026-10-03, curl_cffi 0.16.3 with \u003Ccode>impersonate=\"chrome131\"\u003C\u002Fcode> sent its first request to \u003Ca href=\"https:\u002F\u002Ftls.peet.ws\u002Fapi\u002Fall\" rel=\"noopener noreferrer\">tls.peet.ws\u003C\u002Fa> as \u003Ccode>t13d1516h2_8daaf6152771_02713d6af862\u003C\u002Fcode>. Its second request on the same session, which resumed the TLS session, came back as \u003Ccode>t13d1517h2_8daaf6152771_b0da82dd1658\u003C\u002Fcode>. In short, \u003Ccode>b0da82dd1658\u003C\u002Fcode> means \"a Chromium-style ClientHello, older ALPS codepoint, resumed session\". That covers real older Chrome, Edge, Brave and Opera builds, embedded Chromium, and impersonation libraries pinned to a pre-switch Chrome profile. A JA4 alone cannot tell those apart.\u003C\u002Fp>\n\n\u003Ch2>Reference table: real JA4 values by client\u003C\u002Fh2>\n\u003Cp>The values below come from two sources only: FoxIO's published examples and mapping file, and our own captures against tls.peet.ws on 2026-10-03 (versions stated). Library values depend heavily on the TLS backend and its build options, so treat the library rows as examples, not universal constants. We left out rows we could not verify, including a real Firefox and Safari capture of our own.\u003C\u002Fp>\n\u003Ctable>\n\u003Cthead>\u003Ctr>\u003Cth>Client\u003C\u002Fth>\u003Cth>JA4\u003C\u002Fth>\u003Cth>Source\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\n\u003Ctbody>\n\u003Ctr>\u003Ctd>Headless Chrome 148 (Linux), new connection\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d1516h2_8daaf6152771_d8a2da3f94cd\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture, Playwright Chromium\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Same browser, resumed session\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d1517h2_8daaf6152771_b6f405a00624\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Chromium, older ALPS codepoint, new \u002F resumed\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d1516h2_8daaf6152771_02713d6af862\u003C\u002Fcode> \u002F \u003Ccode>t13d1517h2_8daaf6152771_b0da82dd1658\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>FoxIO mapping CSV; reproduced with curl_cffi \u003Ccode>chrome131\u003C\u002Fcode>\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Chrome over QUIC\u003C\u002Ftd>\u003Ctd>\u003Ccode>q13d0312h3_55b375c5d22e_178839b6cec1\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4#examples\" rel=\"noopener noreferrer\">FoxIO README\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Mozilla Firefox (version not stated)\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d1715h2_5b57614c22b0_7121afd63204\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>FoxIO mapping CSV\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Safari (version not stated)\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d2014h2_a09f3c656075_14788d8d241b\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>FoxIO mapping CSV\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>curl 8.5.0 \u002F OpenSSL 3.0.13 (Ubuntu)\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d3112h2_e8f1e7e78f70_b26ce05bbdd6\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture; identical on tls.peet.ws, BrowserLeaks and Scrapfly\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Python requests 2.31 \u002F urllib3 2.0.7, OpenSSL 3.0.13\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d3112h1_e8f1e7e78f70_b26ce05bbdd6\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Go 1.24.1 net\u002Fhttp (default client)\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d1311h2_f57a46bbacb6_e7c285222651\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>curl_cffi 0.16.3, \u003Ccode>impersonate=\"chrome146\"\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d1516h2_8daaf6152771_d8a2da3f94cd\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>curl_cffi 0.16.3, \u003Ccode>impersonate=\"firefox147\"\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d1717h2_5b57614c22b0_3cbfd9057e0d\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>curl_cffi 0.16.3, \u003Ccode>impersonate=\"safari2601\"\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>\u003Ccode>t13d2013h2_a09f3c656075_7f0f34a4126d\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Our capture\u003C\u002Ftd>\u003C\u002Ftr>\n\u003C\u002Ftbody>\n\u003C\u002Ftable>\n\u003Cp>The patterns matter more than the individual rows:\u003C\u002Fp>\n\u003Cul>\n\u003Cli>\u003Cstrong>The c section often fingerprints the TLS library, not the application.\u003C\u002Fstrong> curl and Python on the same OpenSSL 3 build share \u003Ccode>b26ce05bbdd6\u003C\u002Fcode>. FoxIO's mapping file has a Python entry with a different prefix but the same c. Go clients in FoxIO's file share \u003Ccode>e7c285222651\u003C\u002Fcode> with our Go 1.24 capture.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>requests shows \u003Ccode>h1\u003C\u002Fcode>.\u003C\u002Fstrong> urllib3 2.x advertises only \u003Ccode>http\u002F1.1\u003C\u002Fcode> in ALPN, so the a section tells a server it is not talking to a browser before any header is read.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>A plain library is easy to spot by its counts alone.\u003C\u002Fstrong> 31 ciphers (OpenSSL default) or 13 ciphers with 11 extensions (Go) looks nothing like the 15\u002F16 pattern every current Chromium build sends.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Resumption changes c for every browser.\u003C\u002Fstrong> Firefox goes from 17 to 18 extensions and Safari from 13 to 14 on resumed sessions, in our curl_cffi captures. A detection rule that only allow-lists the fresh-connection value will misfire on real returning visitors.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cp>Expect the Chrome row to keep moving. curl_cffi's newest \u003Ccode>chrome150\u003C\u002Fcode> profile adds three signature algorithms, \u003Ccode>0904\u003C\u002Fcode>, \u003Ccode>0905\u003C\u002Fcode> and \u003Ccode>0906\u003C\u002Fcode>. Those are the \u003Ca href=\"https:\u002F\u002Fdatatracker.ietf.org\u002Fdoc\u002Fdraft-ietf-tls-mldsa\u002F\" rel=\"noopener noreferrer\">ML-DSA codepoints\u003C\u002Fa> from the IETF draft, and they turn c into \u003Ccode>806a8c22fdea\u003C\u002Fcode>. We could not confirm that against a stable Chrome 150 build, so treat it as a preview, not a reference value.\u003C\u002Fp>\n\n\u003Ch2>Why JA4 sorts ciphers and extensions\u003C\u002Fh2>\n\u003Cp>JA4 sorts ciphers and extensions because Chrome deliberately randomises its extension order, and an order-sensitive fingerprint (JA3) stopped being stable. Chrome shipped \u003Ca href=\"https:\u002F\u002Fgroups.google.com\u002Fa\u002Fchromium.org\u002Fg\u002Fblink-dev\u002Fc\u002FbYZK81WxYBo\u002Fm\u002FlKLrZ_P2BwAJ\" rel=\"noopener noreferrer\">TLS ClientHello extension permutation\u003C\u002Fa> around Chrome 110 in early 2023. The goal was to stop servers and middleboxes from depending on a fixed layout. \u003Ca href=\"https:\u002F\u002Fwww.fastly.com\u002Fblog\u002Fa-first-look-at-chromes-tls-clienthello-permutation-in-the-wild\" rel=\"noopener noreferrer\">Fastly measured the effect\u003C\u002Fa>: since JA3 hashes extensions in wire order, one Chrome install produces a different JA3 on practically every connection.\u003C\u002Fp>\n\u003Cp>JA4 solves this by sorting before hashing. FoxIO published it as part of the JA4+ suite in September 2023 (\u003Ca href=\"https:\u002F\u002Fblog.apnic.net\u002F2023\u002F11\u002F22\u002Fja4-network-fingerprinting\u002F\" rel=\"noopener noreferrer\">APNIC republication\u003C\u002Fa>). It is now exposed by \u003Ca href=\"https:\u002F\u002Fdevelopers.cloudflare.com\u002Fbots\u002Fadditional-configurations\u002Fja3-ja4-fingerprint\u002F\" rel=\"noopener noreferrer\">Cloudflare Bot Management\u003C\u002Fa>, AWS CloudFront and WAF, Google Cloud Armor, Fastly, Akamai and others listed in the \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4#tools-that-support-ja4\" rel=\"noopener noreferrer\">FoxIO README\u003C\u002Fa>.\u003C\u002Fp>\n\u003Cp>For scrapers this cuts both ways. Randomising your extension order no longer changes your fingerprint, so that trick is dead. What still matters is the \u003Cem>set\u003C\u002Fem> of ciphers, the \u003Cem>set\u003C\u002Fem> of extensions and the \u003Cem>order\u003C\u002Fem> of signature algorithms. In practice that means using the TLS stack a real browser uses, or a library that copies one closely. If you need the original-order view (for example to compare against JA3), the spec defines \u003Ccode>JA4_o\u003C\u002Fcode> and the raw variants \u003Ccode>JA4_r\u003C\u002Fcode> \u002F \u003Ccode>JA4_ro\u003C\u002Fcode>. BrowserLeaks returns all four. Our \u003Ca href=\"\u002Fblog\u002Ftls-fingerprinting-explained\">TLS fingerprinting explainer\u003C\u002Fa> covers JA3 in more depth.\u003C\u002Fp>\n\n\u003Ch2>The JA4+ family: JA4S, JA4H, JA4X, JA4T\u003C\u002Fh2>\n\u003Cp>JA4 is one method in a larger suite. The others look at the server, the HTTP layer, certificates and TCP, and they come under a different licence.\u003C\u002Fp>\n\u003Ctable>\n\u003Cthead>\u003Ctr>\u003Cth>Method\u003C\u002Fth>\u003Cth>Fingerprints\u003C\u002Fth>\u003Cth>Format, briefly\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\n\u003Ctbody>\n\u003Ctr>\u003Ctd>JA4S\u003C\u002Ftd>\u003Ctd>Server's ServerHello\u003C\u002Ftd>\u003Ctd>protocol + version + extension count + ALPN, then the chosen cipher, then a hash of the server's extensions\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>JA4H\u003C\u002Ftd>\u003Ctd>HTTP request\u003C\u002Ftd>\u003Ctd>method (\u003Ccode>ge\u003C\u002Fcode>, \u003Ccode>po\u003C\u002Fcode>…), version (\u003Ccode>11\u003C\u002Fcode>, \u003Ccode>20\u003C\u002Fcode>), cookie \u003Ccode>c\u003C\u002Fcode>\u002F\u003Ccode>n\u003C\u002Fcode>, referer \u003Ccode>r\u003C\u002Fcode>\u002F\u003Ccode>n\u003C\u002Fcode>, header count, first 4 chars of Accept-Language; then hashes of header names \u003Cem>in order\u003C\u002Fem>, sorted cookie names, and sorted cookie name=value pairs\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>JA4X\u003C\u002Ftd>\u003Ctd>X.509 certificate\u003C\u002Ftd>\u003Ctd>hashes of issuer RDN OIDs, subject RDN OIDs and extension OIDs (how the cert was built, not its values)\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>JA4T\u003C\u002Ftd>\u003Ctd>TCP SYN\u003C\u002Ftd>\u003Ctd>window size, TCP options in order, MSS, window scale, e.g. \u003Ccode>64240_2-1-3-1-1-4_1460_8\u003C\u002Fcode> for Windows 11\u003C\u002Ftd>\u003C\u002Ftr>\n\u003C\u002Ftbody>\n\u003C\u002Ftable>\n\u003Cp>Formats are taken from FoxIO's \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4\u002Ftree\u002Fmain\u002Ftechnical_details\" rel=\"noopener noreferrer\">technical details diagrams\u003C\u002Fa> and the reference \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4\u002Fblob\u002Fmain\u002Fpython\u002Fja4h.py\" rel=\"noopener noreferrer\">Python implementation\u003C\u002Fa>. JA4H matters most for scraping. Its b section hashes header names \u003Cem>in the order sent\u003C\u002Fem>, so a client that sends Chrome's headers in the wrong order is visible even if every value is right. Our \u003Ca href=\"\u002Fblog\u002Fhttp2-fingerprinting-explained-2026\">HTTP\u002F2 fingerprinting post\u003C\u002Fa> covers the related SETTINGS and pseudo-header order signals. JA4T exposes the OS: a Linux server's SYN does not look like Windows, whatever the User-Agent says.\u003C\u002Fp>\n\u003Ch3>Licensing\u003C\u002Fh3>\n\u003Cp>According to the \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4#licensing\" rel=\"noopener noreferrer\">repository's licensing section\u003C\u002Fa>, JA4 (the TLS client fingerprint) is BSD 3-Clause, like JA3, and FoxIO says it has no patent claims on it. JA4S, JA4L, JA4LS, JA4H, JA4X, JA4SSH, JA4T, JA4TS, JA4TScan, JA4D, JA4D6 and the other \"JA4+\" methods are patent-pending and licensed under the FoxIO License 1.1. That licence allows internal and academic use, but a vendor selling JA4+ fingerprinting as part of a product needs an OEM licence.\u003C\u002Fp>\n\n\u003Ch2>QUIC and HTTP\u002F3: what the q prefix tells you\u003C\u002Fh2>\n\u003Cp>A JA4 starting with \u003Ccode>q\u003C\u002Fcode> was computed from a QUIC Initial packet, meaning the client was setting up HTTP\u002F3. QUIC carries a TLS 1.3 ClientHello inside its CRYPTO frames. The Initial packet's protection keys are derived from the Destination Connection ID the client chooses (\u003Ca href=\"https:\u002F\u002Fwww.rfc-editor.org\u002Frfc\u002Frfc9001#section-5.2\" rel=\"noopener noreferrer\">RFC 9001 §5.2\u003C\u002Fa>), so any on-path observer or edge server can decrypt it and fingerprint the ClientHello exactly as it would over TCP.\u003C\u002Fp>\n\u003Cp>Chrome's QUIC fingerprint looks very different from its TCP one: \u003Ccode>q13d0312h3_55b375c5d22e_178839b6cec1\u003C\u002Fcode> in FoxIO's example. That is 3 ciphers (TLS 1.3 only, since QUIC forbids older versions) and ALPN \u003Ccode>h3\u003C\u002Fcode>. This matters for automation because browsers move to HTTP\u002F3 once a site advertises it, while most HTTP clients never do. Cloudflare's documented \u003Ca href=\"https:\u002F\u002Fdevelopers.cloudflare.com\u002Fbots\u002Fadditional-configurations\u002Fja3-ja4-fingerprint\u002F\" rel=\"noopener noreferrer\">JA4 Signals\u003C\u002Fa> include an \u003Ccode>h2h3_ratio_1h\u003C\u002Fcode> value: the share of traffic for a given JA4 arriving over HTTP\u002F2 versus HTTP\u002F3. A \"Chrome\" TLS fingerprint that never shows up over QUIC on an HTTP\u002F3 site is a statistical outlier. The curl_cffi README states that HTTP\u002F3 fingerprint impersonation has been supported since v0.15.0.\u003C\u002Fp>\n\n\u003Ch2>How to see your own JA4\u003C\u002Fh2>\n\u003Cp>The fastest check is an echo service that reads your ClientHello and returns the fingerprint. These three returned JA4 in our tests, and they agreed with each other byte for byte:\u003C\u002Fp>\n\u003Cul>\n\u003Cli>\u003Ca href=\"https:\u002F\u002Ftls.peet.ws\u002Fapi\u002Fall\" rel=\"noopener noreferrer\">tls.peet.ws\u002Fapi\u002Fall\u003C\u002Fa>: JSON with \u003Ccode>tls.ja4\u003C\u002Fcode>, \u003Ccode>tls.ja4_r\u003C\u002Fcode>, JA3, the HTTP\u002F2 Akamai fingerprint and \u003Ccode>http_version\u003C\u002Fcode>.\u003C\u002Fli>\n\u003Cli>\u003Ca href=\"https:\u002F\u002Ftls.browserleaks.com\u002Fjson\" rel=\"noopener noreferrer\">tls.browserleaks.com\u002Fjson\u003C\u002Fa>: JSON with \u003Ccode>ja4\u003C\u002Fcode>, \u003Ccode>ja4_r\u003C\u002Fcode>, \u003Ccode>ja4_o\u003C\u002Fcode>, \u003Ccode>ja4_ro\u003C\u002Fcode>.\u003C\u002Fli>\n\u003Cli>\u003Ca href=\"https:\u002F\u002Fscrapfly.io\u002Fweb-scraping-tools\u002Fja3-fingerprint\" rel=\"noopener noreferrer\">Scrapfly's JA3\u002FJA4 tool\u003C\u002Fa>, whose API at \u003Ccode>tools.scrapfly.io\u002Fapi\u002Ffp\u002Fja3\u003C\u002Fcode> returns \u003Ccode>ja4\u003C\u002Fcode> and \u003Ccode>ja4_r\u003C\u002Fcode>.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cpre>\u003Ccode class=\"language-bash\">curl -s https:\u002F\u002Ftls.peet.ws\u002Fapi\u002Fall | python3 -c \\\n  \"import sys, json; d = json.load(sys.stdin); print(d['http_version'], d['tls']['ja4'])\"\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>Run the same check \u003Cem>through your proxy\u003C\u002Fem>, not just from your laptop. A forward proxy that tunnels with CONNECT passes your ClientHello through untouched, but a TLS-intercepting proxy, corporate middlebox or some SDK layers will substitute their own. For example, with a ProxyHat residential exit:\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-bash\">curl -s -x http:\u002F\u002FUSERNAME:PASSWORD@gate.proxyhat.com:8080 https:\u002F\u002Ftls.peet.ws\u002Fapi\u002Fall\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Ch3>Wireshark and tshark\u003C\u002Fh3>\n\u003Cp>Wireshark has computed JA4 natively since 4.2.0, in the display-filter field \u003Ccode>tls.handshake.ja4\u003C\u002Fcode> (and \u003Ccode>tls.handshake.ja4_r\u003C\u002Fcode>), per the \u003Ca href=\"https:\u002F\u002Fwww.wireshark.org\u002Fdocs\u002Fdfref\u002Ft\u002Ftls.html\" rel=\"noopener noreferrer\">TLS field reference\u003C\u002Fa>. FoxIO's \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FFoxIO-LLC\u002Fja4\u002Fblob\u002Fmain\u002Fwireshark\u002FREADME.md\" rel=\"noopener noreferrer\">JA4+ plugin\u003C\u002Fa> (Wireshark 4.4.0 or later) adds the rest of the suite under \u003Ccode>ja4.*\u003C\u002Fcode>, such as \u003Ccode>ja4.ja4s\u003C\u002Fcode>, \u003Ccode>ja4.ja4h\u003C\u002Fcode> and \u003Ccode>ja4.ja4t\u003C\u002Fcode>. To pull fingerprints out of a capture:\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-bash\">tshark -r capture.pcapng -Y \"tls.handshake.ja4\" -T fields \\\n  -e ip.dst -e tls.handshake.extensions_server_name -e tls.handshake.ja4\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>Capture on a machine where you control the client. To fingerprint HTTPS traffic from a headless browser, run it locally and capture on the loopback or outbound interface. The ClientHello is unencrypted, so you do not need TLS keys for JA4. You do need them for JA4H on HTTPS.\u003C\u002Fp>\n\n\u003Ch2>What this means in practice: headers must match the handshake\u003C\u002Fh2>\n\u003Cp>Changing the User-Agent does not change your JA4. A Python script that sends a Chrome User-Agent still presents \u003Ccode>t13d3112h1_e8f1e7e78f70_b26ce05bbdd6\u003C\u002Fcode> in the handshake, before a single header is read. That mismatch (claims Chrome, negotiates like OpenSSL, asks for HTTP\u002F1.1) is one of the cheapest checks an edge can run. It is also exactly what JA4 databases are built to catch.\u003C\u002Fp>\n\u003Cp>There are two mainstream fixes. Both have limits.\u003C\u002Fp>\n\u003Cul>\n\u003Cli>curl_cffi (Python bindings to curl-impersonate) builds the ClientHello and HTTP\u002F2 frames to match a named browser profile: \u003Ccode>requests.get(url, impersonate=\"chrome\")\u003C\u002Fcode>. In our captures its current Chrome profiles produced the same JA4 as a real headless Chrome 148. Our \u003Ca href=\"\u002Fblog\u002Ftls-impersonation-curl-cffi-guide\">curl_cffi guide\u003C\u002Fa> covers sessions and proxies.\u003C\u002Fli>\n\u003Cli>uTLS for Go replaces \u003Ccode>crypto\u002Ftls\u003C\u002Fcode>'s ClientHello with a \"parrot\" such as \u003Ccode>tls.HelloChrome_Auto\u003C\u002Fcode> through \u003Ccode>tls.UClient(conn, &amp;config, helloID)\u003C\u002Fcode>. Its \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Frefraction-networking\u002Futls\" rel=\"noopener noreferrer\">README\u003C\u002Fa> is explicit that parroting can be imperfect and does not extend beyond the ClientHello. Our \u003Ca href=\"\u002Fblog\u002Ffixing-go-net-http-tls-fingerprint\">Go uTLS walkthrough\u003C\u002Fa> shows how to wire it into \u003Ccode>net\u002Fhttp\u003C\u002Fcode>.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch3>What impersonation does not solve\u003C\u002Fh3>\n\u003Cp>A matching JA4 only gets you past the first filter.\u003C\u002Fp>\n\u003Cul>\n\u003Cli>The JA4 must agree with the claimed browser \u003Cem>version\u003C\u002Fem>: a Chrome 148 User-Agent with a pre-switch ALPS codepoint (\u003Ccode>b0da82dd1658\u003C\u002Fcode>) is its own small mismatch.\u003C\u002Fli>\n\u003Cli>JA4H header order must match too.\u003C\u002Fli>\n\u003Cli>JA4T and IP data reveal the host OS and network.\u003C\u002Fli>\n\u003Cli>JavaScript challenges never run in an HTTP client.\u003C\u002Fli>\n\u003Cli>Behaviour (request rate, navigation pattern) is judged separately.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cp>Pick the profile that matches the User-Agent you send, keep sessions alive so resumption looks normal, and keep the exit IP plausible for the traffic. A residential exit fixes the network-reputation layer, not the TLS layer. Check the fingerprint through the proxy, as shown above.\u003C\u002Fp>\n\u003Cp>Finally, the reason to understand fingerprints is to make legitimate automation honest and predictable, not to get past a site that has decided to refuse automated access. Respect robots.txt and terms of service where they apply, and remember that scraping personal data brings legal obligations a perfect JA4 does nothing about.\u003C\u002Fp>\n","JA4 Fingerprint Decoded: What b0da82dd1658 Means | ProxyHat","Decode any JA4 fingerprint field by field: what t13d1516h2 and b0da82dd1658 mean, real JA4 values for Chrome, curl, Python and Go, and how to check yours.","\u002Fblog\u002Fja4-fingerprint-decoded",null,"JA4 Fingerprint Decoded: What b0da82dd1658 Means","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fja4-fingerprint-decoded-og.webp",[18,21,24,27,30],{"question":19,"answer":20},"What is b0da82dd1658 in a JA4 fingerprint?","It is the third (c) section of the JA4: a truncated SHA-256 of the sorted extension list plus signature algorithms. It matches a Chromium-style ClientHello that includes pre_shared_key (a resumed TLS 1.3 session) and the original ALPS codepoint 0x4469. FoxIO's mapping file labels t13d1517h2_8daaf6152771_b0da82dd1658 as Chromium Browser.",{"question":22,"answer":23},"What does t13d1516h2 mean in JA4?","t means TLS over TCP, 13 means TLS 1.3, d means an SNI hostname was sent, 15 is the number of cipher suites, 16 is the number of extensions (GREASE excluded, SNI and ALPN included), and h2 is the first and last character of the first ALPN value, i.e. HTTP\u002F2.",{"question":25,"answer":26},"What is the difference between JA3 and JA4?","JA3 is one MD5 hash of the ClientHello fields in wire order, so Chrome's extension-order randomisation gives it a new JA3 on almost every connection. JA4 sorts ciphers and extensions before hashing, splits the result into a readable prefix and two hashes, and covers QUIC. JA4 is BSD-licensed like JA3; the wider JA4+ suite is under the FoxIO License.",{"question":28,"answer":29},"How can I check my own JA4 fingerprint?","Request tls.peet.ws\u002Fapi\u002Fall, tls.browserleaks.com\u002Fjson or Scrapfly's JA3\u002FJA4 tool from the client you want to test; each returns the JA4 it saw. For packet captures, Wireshark 4.2+ exposes the tls.handshake.ja4 field, which tshark can print with -T fields.",{"question":31,"answer":32},"Does changing the User-Agent change my JA4?","No. JA4 is computed from the TLS ClientHello, which is sent before any HTTP header. A Python or Go client with a Chrome User-Agent still presents its own library's JA4, which is an easy mismatch for a server to spot. Only a different TLS stack, such as curl_cffi or uTLS, changes it.",[34,35,36,37,38],"A JA4 is a_b_c: readable protocol metadata, a hash of sorted ciphers, and a hash of sorted extensions plus signature algorithms.","b0da82dd1658 is a Chromium c section: resumed TLS 1.3 session (pre_shared_key) with the old ALPS codepoint 0x4469.","The same browser has two JA4s: session resumption adds pre_shared_key, bumping the extension count and changing c.","Library JA4s mostly reflect the TLS backend: curl and Python on OpenSSL 3 share the same c hash.","Spoofing headers without matching TLS creates a mismatch; curl_cffi and uTLS fix the handshake, not JavaScript or behaviour.",false,"2026-10-06T09:00:00+00:00","2026-10-06T09:00:20+00:00","2026-10-03T09:31:26+00:00","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fja4-fingerprint-decoded-master.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fja4-fingerprint-decoded-card.webp",13,{"name":47,"slug":48,"bio":49,"avatar_url":14,"website_url":14,"social_links":14},"ProxyHat Team","proxyhat-team","The ProxyHat editorial team covers proxies, web scraping, and data extraction.",[51],{"id":52,"slug":53,"name":54,"description":14,"meta_title":55,"meta_description":56,"parent_id":14},6,"anti-bot-security","Anti-Bot & Security","Anti-Bot & Security Guides | ProxyHat Blog","Technical deep dives into anti-bot systems, detection evasion, and proxy security.",[58,61,65],{"id":52,"slug":59,"name":60},"anti-detection","Anti-Detection",{"id":62,"slug":63,"name":64},16,"anti-bot","Anti-Bot",{"id":66,"slug":67,"name":68},20,"tls-fingerprinting","TLS Fingerprinting",{"ko":70,"de":71,"tr":72,"en":13,"ru":73,"zh":74,"x-default":13},"\u002Fko\u002Fblog\u002Fja4-fingerprint-haeseok","\u002Fde\u002Fblog\u002Fja4-fingerprint-entschluesselt","\u002Ftr\u002Fblog\u002Fja4-parmak-izi-cozumleme","\u002Fru\u002Fblog\u002Fja4-otpechatok-rasshifrovka","\u002Fcn\u002Fblog\u002Fja4-zhiwen-jiexi",{"ko":70,"de":71,"tr":72,"en":13,"ru":73,"zh":74},52,[78,98,115,140],{"id":79,"slug":80,"title":81,"excerpt":82,"featured_image_url":83,"featured_image_alt":14,"card_image_url":84,"og_image_url":85,"published_at":86,"updated_at":87,"content_updated_at":88,"reading_time_minutes":89,"is_featured":39,"indexable":4,"views":90,"author":91,"categories":92,"tags":94},21,"tls-fingerprinting-explained","TLS Fingerprinting Explained: JA3, JA4, and How to Avoid Detection","Learn how TLS fingerprinting reveals your scraper identity through JA3\u002FJA4 hashes, cipher suite analysis, and HTTP\u002F2 signals — plus mitigation strategies using browser-grade TLS libraries.","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog\u002Ftls-fingerprinting-explained-master.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog\u002Ftls-fingerprinting-explained-card.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog\u002Ftls-fingerprinting-explained-og.webp","2026-01-24T13:57:14+00:00","2026-07-19T03:13:07+00:00","2026-02-25T23:44:11+00:00",9,662,{"name":47,"slug":48,"bio":49,"avatar_url":14,"website_url":14,"social_links":14},[93],{"id":52,"slug":53,"name":54,"description":14,"meta_title":55,"meta_description":56,"parent_id":14},[95,96,97],{"id":62,"slug":63,"name":64},{"id":66,"slug":67,"name":68},{"id":52,"slug":59,"name":60},{"id":99,"slug":100,"title":101,"excerpt":102,"featured_image_url":103,"featured_image_alt":101,"card_image_url":104,"og_image_url":105,"published_at":106,"updated_at":107,"content_updated_at":42,"reading_time_minutes":45,"is_featured":39,"indexable":4,"views":108,"author":109,"categories":110,"tags":112},209,"akamai-bot-manager-cookies-explained","Akamai Bot Manager Cookies Explained: _abck, ak_bmsc, bm_sz, bm_sv","What _abck, ak_bmsc, bm_sz and bm_sv actually are, how long each lives, what the ~0~\u002F~-1~ markers in _abck are believed to mean, and why HTTP-only clients and mid-session IP rotation break them.","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fakamai-bot-manager-cookies-explained-master.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fakamai-bot-manager-cookies-explained-card.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fakamai-bot-manager-cookies-explained-og.webp","2026-10-20T09:00:00+00:00","2026-10-20T09:00:24+00:00",38,{"name":47,"slug":48,"bio":49,"avatar_url":14,"website_url":14,"social_links":14},[111],{"id":52,"slug":53,"name":54,"description":14,"meta_title":55,"meta_description":56,"parent_id":14},[113,114],{"id":62,"slug":63,"name":64},{"id":52,"slug":59,"name":60},{"id":116,"slug":117,"title":118,"excerpt":119,"featured_image_url":120,"featured_image_alt":118,"card_image_url":121,"og_image_url":122,"published_at":123,"updated_at":124,"content_updated_at":42,"reading_time_minutes":125,"is_featured":39,"indexable":4,"views":126,"author":127,"categories":128,"tags":130},217,"stealth-browser-benchmark-datacenter-residential","Stealth Browser Benchmark: Patchright, nodriver, Camoufox (2026)","54 runs, five automation tools, two IP types. Browser-level stealth decided every bot verdict; the residential IP only moved network-side checks, and not always for the better.","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fstealth-browser-benchmark-datacenter-residential-master.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fstealth-browser-benchmark-datacenter-residential-card.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog-covers\u002Fen\u002Fstealth-browser-benchmark-datacenter-residential-og.webp","2026-10-15T09:00:00+00:00","2026-10-15T09:00:27+00:00",15,59,{"name":47,"slug":48,"bio":49,"avatar_url":14,"website_url":14,"social_links":14},[129],{"id":52,"slug":53,"name":54,"description":14,"meta_title":55,"meta_description":56,"parent_id":14},[131,132,135,136],{"id":52,"slug":59,"name":60},{"id":79,"slug":133,"name":134},"headless-browser","Headless Browser",{"id":62,"slug":63,"name":64},{"id":137,"slug":138,"name":139},17,"browser-fingerprinting","Browser Fingerprinting",{"id":141,"slug":142,"title":143,"excerpt":144,"featured_image_url":145,"featured_image_alt":14,"card_image_url":146,"og_image_url":147,"published_at":148,"updated_at":87,"content_updated_at":149,"reading_time_minutes":150,"is_featured":39,"indexable":4,"views":151,"author":152,"categories":153,"tags":155},51,"handling-cloudflare-blocks-guide","Handling Cloudflare Blocks: A White-Hat Guide to Legitimate Access","Learn how Cloudflare detection works and how to access protected sites legitimately using residential proxies, browser-grade TLS, proper request patterns, and ethical scraping practices.","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog\u002Fhandling-cloudflare-blocks-guide-master.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog\u002Fhandling-cloudflare-blocks-guide-card.webp","https:\u002F\u002Fstatic.proxyhat.com\u002Fstorage\u002Fblog\u002Fhandling-cloudflare-blocks-guide-og.webp","2026-02-25T10:14:01+00:00","2026-10-03T08:21:27+00:00",10,415,{"name":47,"slug":48,"bio":49,"avatar_url":14,"website_url":14,"social_links":14},[154],{"id":52,"slug":53,"name":54,"description":14,"meta_title":55,"meta_description":56,"parent_id":14},[156,157],{"id":62,"slug":63,"name":64},{"id":52,"slug":59,"name":60},[]]