diff --git a/src/_posts/2022-01-06-unfinished-projects.md b/src/_posts/2022-01-06-unfinished-projects.md
index 00836b6..36fb238 100644
--- a/src/_posts/2022-01-06-unfinished-projects.md
+++ b/src/_posts/2022-01-06-unfinished-projects.md
@@ -51,7 +51,7 @@ This whole pipeline works flawlessly, but the tooling was never released due to
 
 I actually published one of my failed projects as reference material for other developers.
 
-{{ github(repo="ewpratten/animal-loader") }}
+[https://github.com/ewpratten/animal-loader](https://github.com/ewpratten/animal-loader)
 
 <br>
 
diff --git a/src/_posts/2022-01-19-monocular-blender.md b/src/_posts/2022-01-19-monocular-blender.md
index 065f6d5..2da42b8 100644
--- a/src/_posts/2022-01-19-monocular-blender.md
+++ b/src/_posts/2022-01-19-monocular-blender.md
@@ -65,9 +65,6 @@ Essentially, this toolkit leverages an existing Docker Engine on a host, and pro
 
 Using the toolkit, I threw together a quick project called `midas-depth-solve` that provides a Docker container to run MiDaS through a little [batch-processing wrapper script](https://github.com/Ewpratten/midas-depth-solve/blob/master/solve.py) I wrote. Simply provide a directory full of images in whatever format you'd like along with some configuration flags, and it will spit out each image as a grayscale depth map.
 
-{{ github(repo="ewpratten/midas-depth-solve") }}
-<br>
-
 Information on how to use this container stand-alone yourself can be found in the project README.
 
 An example of an output from MiDaS is shown below. I have boosted the exposure a lot to make it easier to see the depth levels. Generally, depth maps are low-contrast.
diff --git a/src/_posts/2022-04-06-udm-fan.md b/src/_posts/2022-04-06-udm-fan.md
index 3e0a31d..fd27935 100644
--- a/src/_posts/2022-04-06-udm-fan.md
+++ b/src/_posts/2022-04-06-udm-fan.md
@@ -44,7 +44,7 @@ Who needs SSH anyways?
 
 When I discovered this, I decided to write a little program that wraps the SSH commands with a little bit of error checking, and support for auto-discovery of UDM-Pro devices on your network.
 
-{{ github(repo="ewpratten/udm-fan") }}
+[https://github.com/ewpratten/udm-fan](https://github.com/ewpratten/udm-fan)
 
 <br>
 
diff --git a/src/_posts/2022-06-26-rickrolling-the-internet.md b/src/_posts/2022-06-26-rickrolling-the-internet.md
index 5d841ef..1b42a1e 100644
--- a/src/_posts/2022-06-26-rickrolling-the-internet.md
+++ b/src/_posts/2022-06-26-rickrolling-the-internet.md
@@ -57,7 +57,6 @@ I then delegated reverse DNS to the [Hurricane Electric DNS service](https://dns
 
 For this process, I ended up building a little Python tool called `lyrics2ptr` that generates copy-pastable Hurricane Electric settings from a text file full of lyrics.
 
-{{ github(repo="ewpratten/lyrics2ptr") }}
 <br>
 
 The specific command for this was:
@@ -123,10 +122,7 @@ Thus, when `mtr` looks for the third router in line, it'll get the address `2a06
 
 ## Profit!
 
-Well, thats about it. I skipped over some implementation details, but if you'd like to check out the source code for this project, head over to its GitHub page:
-
-{{ github(repo="ewpratten/imaginary-addrs") }}
-<br>
+Well, thats about it. I skipped over some implementation details, but if you'd like to check out the source code for this project, head over to [its GitHub page](https://github.com/ewpratten/imaginary-addrs).
 
 And for the end result: