{"id":318,"date":"2026-10-10T17:56:35","date_gmt":"2026-10-10T15:56:35","guid":{"rendered":"https:\/\/hands-on-security.com\/new-site-asdfbvi\/index.php\/2026\/10\/10\/from-1500-to-under-100-tpm-sniffing\/"},"modified":"2026-10-17T15:26:56","modified_gmt":"2026-10-17T13:26:56","slug":"from-1500-to-under-100-tpm-sniffing","status":"publish","type":"post","link":"https:\/\/hands-on-security.com\/index.php\/2026\/10\/10\/from-1500-to-under-100-tpm-sniffing\/","title":{"rendered":"Sniffing a BitLocker key with an $80 logic analyzer"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">A few weeks ago, we wired up a logic analyzer the size of a postage stamp to the mainboard of a Lenovo L480, started it recording, and switched the laptop on. By the time Windows reached the login screen, the capture held the key that unlocks the machine&#8217;s BitLocker-encrypted disk.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The analyzer was a Sipeed SLogic16U3, and it cost about $80. When we started this work a couple of years ago, our instrument was a $1,500 bench analyzer. We were not hunting for the cheapest thing that could scrape by. You can rig up a capture with a few dollars of parts, and people have, but that was never our interest. We wanted a properly capable, professional-grade instrument, and the technology has simply moved far enough that $80 now buys one. It matches our old $1,500 box, and on the one measurement that matters for this attack, it beats it.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">That is the story worth telling, and it is bigger than a single gadget. The cost of a capable attack rig keeps dropping while what it can do keeps climbing. The budget that once kept this kind of work inside a well-funded lab has mostly evaporated. The takeaway is not that logic analyzers are dangerous. It is that the defenses are what carry the weight now, because a modest budget buys an attacker far more today than it did a few years ago. We own all three analyzers in this post, so what follows is a comparison we can stand behind rather than a reading of the datasheets.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">What we are actually listening to<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Here is the whole attack in one sentence. BitLocker encrypts your disk, the key that does the encrypting lives in a small chip called the TPM, and on a lot of laptops that chip sends the key across an open wire every time the machine boots.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Here it is a little more slowly. In Microsoft&#8217;s default configuration, TPM-only with no startup PIN, the laptop has to unlock itself with nobody present to type anything. So during boot the TPM releases the key material, the volume master key or VMK, to the rest of the system. On many machines the TPM is a separate chip next to the CPU, joined to it by a simple, slow bus called SPI, or the older LPC. If nothing encrypts the traffic on that bus, and in the default setup nothing does, then anything that can touch those wires can read the key as it goes past.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">That is all we are doing. We are not breaking AES, and we are not guessing a password. We are listening in on a conversation the computer has with itself, and writing down the key as it passes. With the VMK in hand, we can derive the key for the volume, read the disk and obtain the BitLocker Recovery Password.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">It does not work on everything, and the exceptions are exactly the defenses:<\/p>\n\n\n\n<ul>\n<li>A <strong>firmware TPM<\/strong>, built into the CPU itself, has no external bus to tap.<\/li>\n<li>Some platforms <strong>encrypt<\/strong> the bus, so a probe sees only ciphertext.<a href=\"#fn1\" class=\"footnote-ref\" id=\"fnref1\" role=\"doc-noteref\" aria-label=\"Footnote 1: HP TPM Guard\"><sup>1<\/sup><\/a><\/li>\n<li><strong>TPM+PIN<\/strong> puts a human secret in front of the key release, so the boot-time conversation no longer gives up the VMK on its own.<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">Three analyzers, one job<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Our three analyzers all have 16 digital channels and a USB 3.0 connection. That is about where the family resemblance ends.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img src=\"https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/saleae-dslogic-sipeed-comparison-cropped.jpg\" alt=\"Saleae Logic Pro 16, DSLogic U3Pro16 and Sipeed SLogic16U3 side by side on the same measurement grid.\" class=\"wp-image-341\" width=\"1200\" height=\"630\" loading=\"lazy\" decoding=\"async\" srcset=\"https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/saleae-dslogic-sipeed-comparison-cropped.jpg 1200w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/saleae-dslogic-sipeed-comparison-cropped-300x158.jpg 300w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/saleae-dslogic-sipeed-comparison-cropped-1024x538.jpg 1024w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/saleae-dslogic-sipeed-comparison-cropped-768x403.jpg 768w\" sizes=\"auto, (max-width: 1200px) 100vw, 1200px\" \/><figcaption>Our three analyzers, shown at the same scale: Saleae, DSLogic and Sipeed.<\/figcaption><\/figure>\n\n\n\n<div class=\"table-scroll hos-capture-table\" role=\"region\" aria-label=\"Logic analyzer comparison\" tabindex=\"0\"><table>\n<thead>\n<tr>\n<th scope=\"col\"><\/th>\n<th scope=\"col\">Saleae Logic Pro 16<\/th>\n<th scope=\"col\">DSLogic U3Pro16<\/th>\n<th scope=\"col\">Sipeed SLogic16U3<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<th scope=\"row\">Price (approx.)<\/th>\n<td>$1,500<\/td>\n<td>$300<\/td>\n<td>$80<\/td>\n<\/tr>\n<tr>\n<th scope=\"row\">Footprint<\/th>\n<td>92 \u00d7 92 mm<\/td>\n<td>79 \u00d7 74 mm<\/td>\n<td>40 \u00d7 40 mm<\/td>\n<\/tr>\n<tr class=\"capture-spi\">\n<th scope=\"row\">SPI capture, 4 channels<\/th>\n<td>500 MS\/s<\/td>\n<td>500 MS\/s<\/td>\n<td>800 MS\/s<\/td>\n<\/tr>\n<tr>\n<th scope=\"row\">LPC capture, 6 channels<\/th>\n<td>500 MS\/s<\/td>\n<td>500 MS\/s<\/td>\n<td>400 MS\/s<\/td>\n<\/tr>\n<tr>\n<th scope=\"row\">Records through boot<\/th>\n<td>streaming<\/td>\n<td>streaming + buffered<\/td>\n<td>streaming<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Prices in US dollars, checked 13 October 2026: <a href=\"https:\/\/www.saleae.com\/products\/logic-pro-16\" target=\"_blank\" rel=\"noopener noreferrer\">Saleae lists $1,499<\/a>, <a href=\"https:\/\/www.dreamsourcelab.com\/product\/dslogic-series\/\" target=\"_blank\" rel=\"noopener noreferrer\">DreamSourceLab $299<\/a>, <a href=\"https:\/\/wiki.sipeed.com\/hardware\/en\/logic_analyzer\/slogic16u3\/Introduction.html\" target=\"_blank\" rel=\"noopener noreferrer\">Sipeed from $69<\/a>. We paid about $80. Shipping, tax and bundles vary.<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The surprise sits in the SPI row. The smallest, cheapest box samples the four signals we care about <em>faster<\/em> than the $1,500 one does. For this particular job, the price tag and the capability point in opposite directions.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">One caveat sits behind those numbers. &#8220;MS\/s&#8221; is megasamples per second, how many times a second the analyzer checks each wire, and the rate a box can hold drops as you turn more channels on. The Sipeed&#8217;s 800 MS\/s is for the four channels SPI needs; LPC needs six, and there it eases back to 400 MS\/s while the other two hold 500, as the table shows. The headline rate always comes with a channel count attached. The full per-channel tables are in the vendor docs if you want them: <a href=\"https:\/\/www.saleae.com\/support\/logic-software\/download-and-installation\/changelog\" target=\"_blank\" rel=\"noopener noreferrer\">Saleae&#8217;s firmware changelog<\/a>, the <a href=\"https:\/\/www.dreamsourcelab.com\/doc\/DSLogic_U3Pro16_Datasheet.pdf\" target=\"_blank\" rel=\"noopener noreferrer\">DSLogic datasheet<\/a> and <a href=\"https:\/\/wiki.sipeed.com\/hardware\/en\/logic_analyzer\/slogic16u3\/Introduction.html\" target=\"_blank\" rel=\"noopener noreferrer\">Sipeed&#8217;s spec page<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Want the cheap one? <a href=\"https:\/\/s.click.aliexpress.com\/e\/_c2RsXxyN\" target=\"_blank\" rel=\"sponsored noopener noreferrer\">SLogic16U3 on AliExpress<\/a>. That is an affiliate link, so we may earn a small commission if you buy through it.<\/em><\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Does $80 really keep up?<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">So the Sipeed samples fast on paper. Does the number mean anything in practice? This is where it helps to be clear about what a sample rate does and does not buy you.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The bus on the board runs at some clock, say 24 MHz for a typical SPI TPM. The sample rate is a different thing entirely: it is how often our analyzer looks at the wire. At 100 MS\/s it looks every 10 nanoseconds. At 800 MS\/s, every 1.25. The more often it looks, the more cleanly it can tell a clock edge apart from a bit flipping on a data line. What the number does <em>not<\/em> mean is that an &#8220;800 MS\/s&#8221; box can read an 800 MHz bus. The sample rate has to beat the bus clock by a comfortable margin, not match it.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">You might reach for the textbook Nyquist rule and sample at twice the clock. But that rule is about reconstructing a smooth analog wave, not about catching the sharp edges and tight timing of a digital bus with a sampling clock of your own. In practice you want more room. <a href=\"https:\/\/www.saleae.com\/support\/logic-software\/capturing-data\/what-sample-rate-is-required\" target=\"_blank\" rel=\"noopener noreferrer\">Saleae suggests at least four times the SPI clock<\/a>; we tend to aim higher, then look at the trace before we trust the decode.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Here is what &#8220;higher&#8221; works out to for us:<\/p>\n\n\n\n<div class=\"table-scroll hos-capture-table\" role=\"region\" aria-label=\"Practical sampling targets\" tabindex=\"0\"><table>\n<thead>\n<tr>\n<th scope=\"col\">Bus<\/th>\n<th scope=\"col\">Clock<\/th>\n<th scope=\"col\">4\u00d7 rule of thumb<\/th>\n<th scope=\"col\">What we aim for<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr class=\"capture-spi\">\n<th scope=\"row\">SPI, 4 signals<\/th>\n<td>10\u201324 MHz<\/td>\n<td>40\u201396 MS\/s<\/td>\n<td>100 MS\/s or more<\/td>\n<\/tr>\n<tr>\n<th scope=\"row\">Faster SPI, 4 signals<\/th>\n<td>66 MHz<\/td>\n<td>264 MS\/s<\/td>\n<td>500 MS\/s or more<\/td>\n<\/tr>\n<tr>\n<th scope=\"row\">LPC, 6 signals<\/th>\n<td>33 MHz<\/td>\n<td>132 MS\/s<\/td>\n<td>250 MS\/s or more<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n\n\n\n<p class=\"wp-block-paragraph\">And the 500-against-800 gap from earlier mostly evaporates. A 24 MHz SPI TPM wants a target around 100 MS\/s, and every analyzer here clears that several times over. The expensive instrument is not the thing that buys you a successful capture. Short probe leads, a solid ground, the right voltage threshold and enough input bandwidth matter at least as much as the raw rate, and not one of those is a line item that scales with the price.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Why we record the whole boot<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">On a laptop we have never seen before, we do not know exactly when the TPM will hand over the key. It might be early, it might be late. So instead of setting a clever trigger and hoping we armed it at the right instant, we start recording before we press power and go looking for the exchange afterward. That is what &#8220;streaming&#8221; means here: the analyzer sends samples to the PC continuously, for as long as we need, rather than filling a small internal buffer and stopping.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The catch is that the PC has to keep up. Four channels at 800 MS\/s produce about 3.2 gigabits, or 400 megabytes, every second, before any overhead. How long you can record then depends on USB throughput, host memory, compression and the software. It is also why, for this job, we reach for a streaming logic analyzer rather than a high-end oscilloscope. An oscilloscope gives you a beautiful picture of a few microseconds, but we need the whole boot, not a sharper snapshot. When there is a voltage level to check or ringing on a line to chase, it comes back out. Different tool, different question.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Making it cheap took a while<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">We are standing on a decade of other people&#8217;s work. The cost of capturing TPM traffic has been falling since well before our $80 session, and it is worth naming who moved it:<\/p>\n\n\n\n<ul>\n<li><strong>2018:<\/strong> Jeremy Boone, <a href=\"https:\/\/www.nccgroup.com\/media\/jtohmrn3\/_nccgroup-tpm-genie.pdf\" target=\"_blank\" rel=\"noopener noreferrer\"><em>TPM Genie<\/em><\/a>. Interposer research on the TPM&#8217;s serial bus (I2C).<\/li>\n<li><strong>2019:<\/strong> Denis Andzakovic, <a href=\"https:\/\/pulsesecurity.co.nz\/articles\/TPM-sniffing\" target=\"_blank\" rel=\"noopener noreferrer\"><em>Extracting BitLocker keys from a TPM<\/em><\/a>. LPC capture with an FPGA and a logic analyzer.<\/li>\n<li><strong>2020:<\/strong> Henri Nurmi, <a href=\"https:\/\/labs.reversec.com\/posts\/2020\/12\/sniff-there-leaks-my-bitlocker-key\" target=\"_blank\" rel=\"noopener noreferrer\"><em>Sniff, there leaks my BitLocker key<\/em><\/a>. Capturing and decoding it over SPI.<\/li>\n<li><strong>2023:<\/strong> us, <a href=\"https:\/\/i.blackhat.com\/docs\/usa\/2023\/BHUS23-Training-Map.pdf\" target=\"_blank\" rel=\"noopener noreferrer\"><em>Defeating Microsoft&#8217;s Default BitLocker Implementation<\/em><\/a>. Our Black Hat USA training, with Popp Schweiz AG.<\/li>\n<li><strong>2024:<\/strong> Thomas Roth (stacksmashing), <a href=\"https:\/\/www.youtube.com\/watch?v=wTl4vEednkQ\" target=\"_blank\" rel=\"noopener noreferrer\"><em>Breaking Bitlocker<\/em><\/a>. LPC capture on a few-dollar RP2040 board.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Roth&#8217;s <a href=\"https:\/\/github.com\/stacksmashing\/pico-tpmsniffer\" target=\"_blank\" rel=\"noopener noreferrer\">Pico TPMSniffer<\/a> is about the cheapest way there is to grab LPC traffic, and it is a lovely piece of work. We still reach for a general-purpose analyzer in research and teaching, for one reason: we want the raw waveform sitting next to the decoded bytes. When a capture comes up empty, we have to know whether the target genuinely resisted the attack or whether we simply missed part of the exchange. Only the waveform tells us which.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">The box is the easy part<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Honestly, the analyzer is the least interesting thing on the bench. The fiddly work is physical: finding the right pins, tacking on fine wires, getting a clean ground. After that comes the real work, turning a few hundred megabytes of samples back into a key.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img src=\"https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/slogic16u3-l480-spi-workbench.webp\" alt=\"Sipeed SLogic16U3 connected through a TPM adapter to an open Lenovo L480, beside a host computer running PulseView.\" class=\"wp-image-337\" width=\"1920\" height=\"1435\" loading=\"lazy\" decoding=\"async\" srcset=\"https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/slogic16u3-l480-spi-workbench.webp 1920w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/slogic16u3-l480-spi-workbench-300x224.webp 300w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/slogic16u3-l480-spi-workbench-1024x765.webp 1024w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/slogic16u3-l480-spi-workbench-768x574.webp 768w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/slogic16u3-l480-spi-workbench-1536x1148.webp 1536w\" sizes=\"auto, (max-width: 1920px) 100vw, 1920px\" \/><figcaption>The SLogic16U3 connected to the Lenovo L480&#8217;s SPI TPM bus through our adapter.<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">We used to export the decoder&#8217;s output and recover the key in a separate step. These days we have pulled more of that into PulseView with our own decoders, including our new &#8220;TPM over LPC (Hands-On Security)&#8221; decoder we hand to people in our classes, so the decoded key material sits right beside the signals that produced it. That makes a wrong result much easier to trace back, and much harder to believe by accident.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img src=\"https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder.png\" alt=\"PulseView showing four SPI signals and the Hands-On Security TPM over SPI decoder, with VMK annotations below the captured traffic.\" class=\"wp-image-338\" width=\"2360\" height=\"1158\" loading=\"lazy\" decoding=\"async\" srcset=\"https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder.png 2360w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder-300x147.png 300w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder-1024x502.png 1024w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder-768x377.png 768w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder-1536x754.png 1536w, https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder-2048x1005.png 2048w\" sizes=\"auto, (max-width: 2360px) 100vw, 2360px\" \/><figcaption>Our TPM over SPI decoder in PulseView, with the captured signals and VMK annotations. <a href=\"https:\/\/hands-on-security.com\/wp-content\/uploads\/2026\/10\/pulseview-spi-tpm-decoder.png\">View full-size capture<\/a>.<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">We run all of this on hardware we are authorized to touch, in training and in red-team and penetration-testing work. It is a well-documented technique with a long paper trail; Compass Security published <a href=\"https:\/\/blog.compass-security.com\/2024\/02\/microsoft-bitlocker-bypasses-are-practical\/\" target=\"_blank\" rel=\"noopener noreferrer\">an independent account<\/a> of the hands-on class. And the point of demonstrating it is the defenses, not the trick. A firmware TPM, an encrypted bus, or a startup PIN each shut this particular attack down. The SLogic16U3 joins our class bench for the first time in October 2026.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">We have also backed Sipeed&#8217;s next one, the <a href=\"https:\/\/www.kickstarter.com\/projects\/sipeed\/slogic32u3-the-worlds-first-10gbps-usb32-logic-analyzer\" target=\"_blank\" rel=\"noopener noreferrer\">SLogic32U3<\/a>: 32 channels, up to 1.4 GS\/s on four of them, which should be useful for other hardware research. We have not laid hands on it yet, and will write it up when we have. It is telling, though, that Sipeed now <a href=\"https:\/\/en.wiki.sipeed.com\/hardware\/en\/logic_analyzer\/slogic_agent\/index.html\" target=\"_blank\" rel=\"noopener noreferrer\">shows TPM SPI decoding<\/a> in its own product material. A few years ago this was niche research. Now it is a selling point.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">We work through the attack and its limits on real hardware in <a href=\"https:\/\/hands-on-security.com\/index.php\/training\/breaking-bitlocker\/\">Breaking BitLocker<\/a>.<\/p>\n\n\n\n<section id=\"footnotes\" class=\"wp-block-footnotes\" role=\"doc-endnotes\" aria-label=\"Footnote\">\n<hr>\n<ol>\n<li id=\"fn1\"><p><a href=\"https:\/\/threatresearch.ext.hp.com\/tpm-guard-advancing-endpoint-protection-against-physical-attacks\/\" target=\"_blank\" rel=\"noopener noreferrer\">HP&#8217;s TPM Guard<\/a> uses SPDM to encrypt and authenticate CPU\u2013TPM communication on supported PCs, keeping key material out of a passive bus capture.<a href=\"#fnref1\" class=\"footnote-back\" role=\"doc-backlink\" aria-label=\"Return to TPM Guard reference\"> Back to text<\/a><\/p><\/li>\n<\/ol>\n<\/section>\n\n","protected":false},"excerpt":{"rendered":"<p>A couple of years ago, this work needed a $1,500 bench instrument. Today an $80 one does the same job, and sometimes better. The budget for a serious hardware attack keeps shrinking, which is the real reason to care about the defenses.<\/p>\n","protected":false},"author":1,"featured_media":343,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"hos_description":"We recovered BitLocker key material with an $80 logic analyzer. Comparing SPI and LPC sampling rates, continuous capture and our PulseView workflow.","hos_image":"","hos_social_image":341,"hos_byline":"Pascal Gujer & Joel Frei","hos_author_ids":"136,137","hos_author_type":"Person","hos_author_url":"","hos_related_course":0,"hos_references":"","footnotes":""},"categories":[5],"tags":[6,10,14,17,16],"class_list":["post-318","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-system-security","tag-bitlocker","tag-hardware","tag-logic-analyzers","tag-spi","tag-tpm"],"_links":{"self":[{"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/posts\/318","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/comments?post=318"}],"version-history":[{"count":9,"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/posts\/318\/revisions"}],"predecessor-version":[{"id":342,"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/posts\/318\/revisions\/342"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/media\/343"}],"wp:attachment":[{"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/media?parent=318"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/categories?post=318"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hands-on-security.com\/index.php\/wp-json\/wp\/v2\/tags?post=318"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}