Merge pull request #3144 from madeline-underwood/perfor

Perfor reviewed

Author: pareenaverma

content/learning-paths/servers-and-cloud-computing/performix-mcp-agent/1-overview.md

@@ -1,16 +1,20 @@
 ---
-title: Understand Arm Performix in the Arm MCP Server
+title: Understand AI-driven profiling with Arm Performix MCP
 weight: 2
 
 ### FIXED, DO NOT MODIFY
 layout: learningpathall
 ---
 
+## Why AI-driven profiling with Arm Performix
+
+The Arm MCP Server exposes Arm Performix as a first-class tool that AI coding assistants can invoke directly. Rather than switching between your IDE and the Performix GUI to analyze results and then back again to apply code changes, an AI agent can orchestrate the entire profiling pipeline — configuring the recipe, launching the collection run, retrieving hotspot data, and proposing optimizations — all in a single agentic workflow.
+
 ## What is the Arm Performix tool in the MCP Server?
 
-The Arm MCP Server exposes Arm Performix as a first-class tool that AI coding assistants can invoke directly. Arm Performix is a performance profiling tool that simplifies the workflow of collecting CPU samples, building flame graphs, and identifying the functions that dominate application runtime. When integrated into the MCP server, it lets an AI agent orchestrate the entire profiling pipeline — configuring the recipe, launching the collection run, and retrieving the resulting hotspot data — without manual interaction with the Performix engine. 
+Arm Performix is a performance profiling tool that simplifies the workflow of collecting CPU samples, building flame graphs, and identifying the functions that dominate application runtime. When integrated into the MCP server, it lets an AI agent orchestrate the entire profiling pipeline — configuring the recipe, launching the collection run, and retrieving the resulting hotspot data — without manual interaction with the Performix engine.
 
-This removes a context switch for profiling. Rather than switching between your IDE and the Performix GUI to analyze results and then back again to apply code changes, an AI agent can do all of this for you in a single agentic workflow.
+Context switching for profiling is removed. Rather than switching between your IDE and the Performix GUI to analyze results and then back again to apply code changes, an AI agent can do all of this for you in a single agentic workflow.
 
 ## How the MCP tool works
 
@@ -32,7 +36,7 @@ You can ask your AI assistant direct questions and it will invoke the `apx_recip
 Run the Code Hotspots recipe on /home/ec2-user/Mandelbrot-Example/build/mandelbrot_single_thread_debug and tell me which functions are the hottest
 ```
 
-This is useful for quick, exploratory checks. It works well when you already know the binary path and just want a fast hotspot summary before committing to deeper analysis.
+Direct chat is useful for quick, exploratory checks. It works well when you already know the binary path and just want a fast hotspot summary before committing to deeper analysis.
 
 ### Prompt files
 

content/learning-paths/servers-and-cloud-computing/performix-mcp-agent/2-setup.md

@@ -1,20 +1,24 @@
 ---
-title: Build the example application and configure the target
+title: Build the Mandelbrot example on Arm Neoverse
 weight: 3
 
 ### FIXED, DO NOT MODIFY
 layout: learningpathall
 ---
 
+## Prepare the profiling target
+
+In this section, you'll build the Mandelbrot C++ application on your remote Arm server and confirm that Arm Performix can reach the target.
+
 ## About the example application
 
 This Learning Path profiles the same Mandelbrot C++ application used in the [Find code hotspots with Arm Performix](/learning-paths/servers-and-cloud-computing/cpu_hotspot_performix/) Learning Path. It generates a 1920×1080 bitmap of the Mandelbrot set by iterating a simple recurrence for each pixel and is compute-heavy enough to produce clear profiling signal without requiring a long-running workload. The single-threaded build is intentionally unoptimized so that the hotspot analysis surfaces a meaningful target for improvement.
 
-You don't need to understand the Mandelbrot algorithm to follow this Learning Path. It's used here as a convenient, reproducible benchmark for profiling on Arm Neoverse.
+You don't need to understand the Mandelbrot algorithm to follow this Learning Path.
 
 ## Connect to your Arm target
 
-This Learning Path targets an AWS Graviton3 metal instance (`m7g.metal`) with 64 Neoverse V1 cores. Any Arm Linux server with multiple cores works for the parallelization step, but a metal instance gives you direct access to all hardware threads without the overhead of virtualization.
+For profiling, the Learning Path targets an AWS Graviton3 metal instance (`m7g.metal`) with 64 Neoverse V1 cores. Any Arm Linux server with multiple cores works for the parallelization step, but a metal instance gives you direct access to all hardware threads without the overhead of virtualization.
 
 You connect to the remote target via SSH through the Arm MCP Server. The `apx_recipe_run` tool accepts the target host IP address and SSH username directly as parameters, so there is no separate target configuration step required. Ensure your remote Arm server is reachable over SSH from the machine running your AI coding assistant, and follow the [Configure your MCP client](https://github.com/arm/mcp?tab=readme-ov-file#2-configure-your-mcp-client) instructions in the Arm MCP Server repository before continuing.
 
@@ -36,7 +40,7 @@ cd Mandelbrot-Example
 make single_thread DEBUG=1
 ```
 
-This produces the binary at `./build/mandelbrot_single_thread_debug`. Confirm it exists before continuing:
+The command produces the binary at `./build/mandelbrot_single_thread_debug`. Confirm it exists before continuing:
 
 ```bash
 ls -lh build/mandelbrot_single_thread_debug

content/learning-paths/servers-and-cloud-computing/performix-mcp-agent/3-run-hotspot.md

@@ -1,15 +1,19 @@
 ---
-title: Run the Code Hotspots recipe using a GitHub Copilot agent
+title: Run Code Hotspots with an AI agent
 weight: 4
 
 ### FIXED, DO NOT MODIFY
 layout: learningpathall
 ---
 
+## Execute profiling through the Arm MCP Server
+
+In this section, you'll use a GitHub Copilot prompt file to drive the Code Hotspots recipe through the Arm MCP Server. The agent confirms your target details, runs the recipe autonomously, and returns structured profiling results.
+
 ## Use the Arm MCP performance-beginner prompt file
 
 {{% notice Note %}}
-This section uses Visual Studio Code with GitHub Copilot. If you prefer a different AI assistant, see [Configure other AI agents](#configure-other-ai-agents) at the end of this section for equivalent configurations for Kiro and OpenAI Codex.
+The section uses Visual Studio Code with GitHub Copilot. If you prefer a different AI assistant, see [Configure other AI agents](#configure-other-ai-agents) at the end of this section for equivalent configurations for Kiro and OpenAI Codex.
 {{% /notice %}}
 
 The Arm MCP Server repository includes a ready-made prompt file called `performance-beginner` that guides an AI agent through the full Code Hotspots workflow: baseline profiling, hotspot identification, targeted code changes, and re-profiling to confirm the improvement. You don't need to write this file yourself, you copy it from the repository.

content/learning-paths/servers-and-cloud-computing/performix-mcp-agent/4-optimize.md

@@ -1,12 +1,12 @@
 ---
-title: Apply optimizations and verify improvements
+title: Optimize code with AI-driven profiling feedback
 weight: 5
 
 ### FIXED, DO NOT MODIFY
 layout: learningpathall
 ---
 
-## Apply the agent's proposed optimizations
+## Apply AI-suggested optimizations
 
 In the previous section, the agent identified three optimization opportunities:
 
@@ -47,7 +47,7 @@ The fix is to replace `std::complex<double>` in `getIterations` with plain `doub
 $$re_{new} = re_z^2 - im_z^2 + re_c$$
 $$im_{new} = 2 \cdot re_z \cdot im_z + im_c$$
 
-This eliminates every `std::complex` method call from the inner loop. If the agent hasn't already proposed this change, use the following prompt to direct it:
+The fix eliminates every `std::complex` method call from the inner loop. If the agent hasn't already proposed this change, use the following prompt to direct it:
 
 ```text
 On the remote server, rewrite the getIterations function in

content/learning-paths/servers-and-cloud-computing/performix-mcp-agent/_index.md

@@ -1,10 +1,6 @@
 ---
 title: Identify code hotspots using Arm Performix through the Arm MCP Server
 
-draft: true
-cascade:
-    draft: true
-    
 description: Learn how to use an AI agent and the Performix tool through the Arm MCP Server to run the Code Hotspots recipe on a C++ application, interpret flame graph results, and apply targeted optimizations on Arm Neoverse.
 
 minutes_to_complete: 30