diff --git a/README.md b/README.md
index 0d0e237..dd73560 100644
--- a/README.md
+++ b/README.md
@@ -49,8 +49,8 @@ I've used a few different code generation stacks in the past, and they're all go
 With the goal of creating something that is both lightweight and easy to use, I've decided on mixing Java and non-Java bits. By doing this, I will be able to get the full power of Java within my reach and delegate commonly occurring text output tasks to a few non-Java bits - from this point on the non-Java bits are called **_generative functions_**. 
 
 Let’s mock-up template structures:
-![Template Mock-up 1](https://chulk.org/code-dev/codegen/src/branch/main/doc/images/image4.jpg "Template mock-up 1")
-![Template Mock-up 2](https://chulk.org/code-dev/codegen/src/branch/main/doc/images/image3.jpg "Template mock-up 2")
+![Template Mock-up 1](./doc/images/image4.jpg "Template mock-up 1")
+![Template Mock-up 2](./doc/images/image3.jpg "Template mock-up 2")
 
 
 The first part is to create models that drive text output; the second half involves file and text related things such as file creation and text output.