📖 Added a walkthrough of how to create a custom rule

.trunk/configs/custom-words.txt

@@ -156,3 +156,4 @@ felixge
 cenkalti
 tmpl
 sdktrace
+uberjar

docs/custom-ruleset-example/quarkus-fabric-client.yaml

@@ -0,0 +1,17 @@
+- ruleID: fabric8-remove-kubernetes-model-00001
+  labels:
+    - konveyor.io/target=quarkus
+  effort: 1
+  category: mandatory
+  description: |
+    The Maven artifact io.fabric8:kubernetes-model has been removed from the project and is no longer published.
+    This artifact was just an aggregator of some of the Kubernetes model artifacts and had no specific purpose.
+    It is no longer published, the io.fabric8:kubernetes-client-api or io.fabric8:kubernetes-openshift-uberjar artifacts should be used instead.
+  message: |
+    The Maven artifact io.fabric8:kubernetes-model has been removed from the project and is no longer published.
+
+    This artifact was just an aggregator of some of the Kubernetes model artifacts and had no specific purpose.
+    It is no longer published, the io.fabric8:kubernetes-client-api or io.fabric8:kubernetes-openshift-uberjar artifacts should be used instead.
+  when:
+    java.dependency:
+      name: io.fabric8.kubernetes-model

docs/custom-ruleset-example/ruleset.yaml

@@ -0,0 +1,4 @@
+name: quarkus-3-18
+description: These rules are created from the 3.18 migration guide.
+labels:
+  - konveyor.io/target=quarkus

docs/custom_ruleset.md

@@ -47,3 +47,79 @@ You can now click `Run Analysis` and your custom ruleset will be used.
 
 > [!WARNING]
 > You will need to make sure that the rules or rulesets have a target or source label that is selected. If the ruleset does not match a selected target or source, it will be filtered out and skipped.
+
+## Creating Rules and Rulesets
+
+### Creating Rules
+
+To create your own rules, first you need to know of a change that must happen to complete a migration. This could be because of a framework or library that you are using (either an internal one or external/open source one). For this section, I am going to create a rule and ruleset for an open source framework but the process should be similar for any custom rule.
+
+#### Finding A Rule
+
+To find a place for a rule, we are going to look at the release notes of the quarkus framework. If we see the [section](https://github.com/quarkusio/quarkus/wiki/Migration-Guide-3.18#kubernetes-client-fabric8) here. If we follow this to the migration guide for this dependency, we can see this [section](https://github.com/fabric8io/kubernetes-client/blob/main/doc/MIGRATION-v7.md#kubernetes-model-artifact-removed-).
+
+> The Maven artifact io.fabric8:kubernetes-model has been removed from the project and is no longer published.
+> This artifact was just an aggregator of some of the Kubernetes model artifacts and had no specific purpose. It is no longer published, the io.fabric8:kubernetes-client-api or io.fabric8:kubernetes-openshift-uberjar artifacts should be used instead.
+
+So for this, we will need to make sure that if the dependency `io.fabric8:kubernetes-model` is being used, that we alert the user, and tell them what they should be using instead.
+
+#### Creating a Rule
+
+Now we need to look up the rules and what the java provider can do. The [rules](https://github.com/konveyor/analyzer-lsp/blob/main/docs/rules.md#rule-metadata) documentation states that we will need to set up the metadata first. In this case, we will use the below yaml for this.
+
+```yaml
+- ruleID: "fabric8-remove-kubernetes-model-00001"
+  labels:
+    - "konveyor.io/target=quarkus"
+  effort: 1
+  category: mandatory
+```
+
+Next, we will add a description. This description is going to be used by the front end to show the user what the issue entails, I think that the migration docs do a good job of describing this, so we can just use what they have.
+
+```yaml
+description: |
+  The Maven artifact io.fabric8:kubernetes-model has been removed from the project and is no longer published.
+
+  This artifact was just an aggregator of some of the Kubernetes model artifacts and had no specific purpose. It is no longer published, the io.fabric8:kubernetes-client-api or io.fabric8:kubernetes-openshift-uberjar artifacts should be used instead."
+```
+
+Next we need to decide what `Action` to take. The [actions](https://github.com/konveyor/analyzer-lsp/blob/main/docs/rules.md#rule-actions) are very useful but it worth noting that for KAI, the only action that will be used, is message.
+
+> [!WARNING]
+> If you use only the tag action, or if you don't set effort, then the rule's violations will not be used by kai.
+
+We need to note, that the message is used by the LLM to generate a fix for the issue. In this case, I still think the migration guide docs do a good job.
+
+> [!NOTE]
+> This process may require some iteration for determining the optimal message for generating the fix that you want for your issue and model.
+
+```yaml
+  message: |
+  	The Maven artifact io.fabric8:kubernetes-model has been removed from the project and is no longer published.
+
+  	This artifact was just an aggregator of some of the Kubernetes model artifacts and had no specific purpose. It is no longer published, the io.fabric8:kubernetes-client-api or io.fabric8:kubernetes-openshift-uberjar artifacts should be used instead."
+```
+
+Now that we have all the information captured in our rule for using it in kai, we need to add when this should be triggered. This is the `when` clause for a rule. We usually call these [conditions](https://github.com/konveyor/analyzer-lsp/blob/main/docs/rules.md#rule-conditions), you can do a lot of different things here based on providers. Today in kai, we only have two providers, the java and builtin providers.
+
+For this issue, we are looking at dependencies, and so we will choose to the use the `java.dependency` [capability](https://github.com/konveyor/analyzer-lsp/blob/main/docs/rules.md#provider-condition).
+
+```yaml
+  when:
+  	java.dependency:
+  		name: io.fabric8.kubernetes-model
+```
+
+Now, we should save this file in a directory for this ruleset, and we will need to create a ruleset.yaml in the same directory.
+
+Creating the [ruleset.yaml](https://github.com/konveyor/analyzer-lsp/blob/main/docs/rules.md#ruleset) is a special file that will help group these rules together to help manage them easier. For this rule we will use this ruleset.
+
+```yaml
+name: quarkus-3-18
+description: These rules are created from the 3.18 migration guide.
+labels:
+  - "konveyor.io/target=quarkus"
+```
+
+This full rule should now be usable, you can see it [here](./custom-ruleset-example), with the ruleset.yaml that would be needed. To use this ruleset, now follow the [adding custom rules](#steps-to-add-custom-rulesets) section to use this ruleset!