Fix dogfooding files

Author: danielholanda

dogfooding/README.md

@@ -1,42 +1,14 @@
-# AMD Skills Dogfooding: `local-ai-use`
+# AMD Skills Dogfooding
 
-The goal of this skill is to teach your AI agent to use image generation, text generation, and text to speech locally.
+Welcome to AMD Skills Dogfooding.
 
+## Requirements
 
-## Step 1 - Understanding which skills are available
+This dogfooding program assumes you already have `claude code` setup on your machine.
 
-* Run `claude "Which skills can you see?" --model sonnet`. You should see a list of skills that should not include anythink related to local LLM usage.
-* Make sure there is no `AFENTS.md` file on your local folder.
+Participatns using other are still encouraged to participate. Just please note that you may have to adapt some of the instructions provided below to your specific agent.
 
-## Step 2 - Enabling claude to see `local-ai-use`
 
-In the future this will be enabled directly through claude's marketplace. For now, we have to manually add it.
+Please choose a skill to get started.
 
-* Clone `https://github.com/amd/skills`
-* Move the `local-ai-use` skill from the repo to `.claude/skills/`
-* Run `claude "Which skills can you see?" --model sonnet`. You should see a list of skills that includes `local-ai-use`.
-
-## Step 3 - Running the skill
-
-Open Claude and run the prompt:
-```
-Learn how to do image generation locally
-```
-
-Followed by
-```
-Generate the image of a cat
-```
-
-Claude should install Lemonade locally on your device and allow you to generate images locally after the first setup run.
-
-## Step 4 - (Optional) Going beyond
-
-The `local-ai-use` can also help you with text to speech and speech to text locally. Simply ask claude for help there.
-
-## Step 5 - (Optional) Try to get things done without AMD Skills
-
-Remove the added skills from `.claude/skills/` and rerun the experiment above. This should lead to a high variance in execution length and token usage.
-* Model being successful after significant token usage.
-* Model providing a knowledge article instead of actually learning how to do it.
-* Model attempting to come up with a custom strategy to generate images locally, resulting in very low-quality assets.
+* [local-ai-use](./local-ai-use.md): Teach your agent how to run image generation locally.

dogfooding/local-ai-use.md

@@ -1,14 +1,42 @@
-# AMD Skills Dogfooding
+# AMD Skills Dogfooding: `local-ai-use`
 
-Welcome to AMD Skills Dogfooding.
+The goal of this skill is to teach your AI agent to use image generation, text generation, and text to speech locally.
 
-## Requirements
 
-This dogfooding program assumes you already have `claude code` setup on your machine.
+## Step 1 - Understanding which skills are available
 
-Participatns using other are still encouraged to participate. Just please note that you may have to adapt some of the instructions provided below to your specific agent.
+* Run `claude "Which skills can you see?" --model sonnet`. You should see a list of skills that should not include anythink related to local LLM usage.
+* Make sure there is no `AFENTS.md` file on your local folder.
 
+## Step 2 - Enabling claude to see `local-ai-use`
 
-Please choose a skill to get started.
+In the future this will be enabled directly through claude's marketplace. For now, we have to manually add it.
 
-* [local-ai-use](./local-ai-use.md): Teach your agent how to run image generation locally.
+* Clone `https://github.com/amd/skills`
+* Move the `local-ai-use` skill from the repo to `.claude/skills/`
+* Run `claude "Which skills can you see?" --model sonnet`. You should see a list of skills that includes `local-ai-use`.
+
+## Step 3 - Running the skill
+
+Open Claude and run the prompt:
+```
+Learn how to do image generation locally
+```
+
+Followed by
+```
+Generate the image of a cat
+```
+
+Claude should install Lemonade locally on your device and allow you to generate images locally after the first setup run.
+
+## Step 4 - (Optional) Going beyond
+
+The `local-ai-use` can also help you with text to speech and speech to text locally. Simply ask claude for help there.
+
+## Step 5 - (Optional) Try to get things done without AMD Skills
+
+Remove the added skills from `.claude/skills/` and rerun the experiment above. This should lead to a high variance in execution length and token usage.
+* Model being successful after significant token usage.
+* Model providing a knowledge article instead of actually learning how to do it.
+* Model attempting to come up with a custom strategy to generate images locally, resulting in very low-quality assets.