From 5f16bc87e3c871ab0663d6f2e3f2abe636b93b33 Mon Sep 17 00:00:00 2001 From: Madeline Underwood Date: Tue, 30 Dec 2025 16:55:50 +0000 Subject: [PATCH 1/5] Add comprehensive learning path for Gemini CLI and Arm development --- .../gemini/_index.md | 61 +++++++ .../gemini/_next-steps.md | 8 + .../gemini/arm-code-analysis.md | 42 +++++ .../gemini/background.md | 29 ++++ .../gemini/configure-context.md | 56 ++++++ .../gemini/install.md | 164 ++++++++++++++++++ .../gemini/integrate.md | 77 ++++++++ .../gemini/optimization.md | 42 +++++ .../gemini/understand-mcp.md | 38 ++++ 9 files changed, 517 insertions(+) create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/_index.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/_next-steps.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/arm-code-analysis.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/background.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/configure-context.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/install.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/integrate.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/optimization.md create mode 100644 content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md new file mode 100644 index 0000000000..3ff6496c73 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md @@ -0,0 +1,61 @@ +--- +title: Analyze and optimize Arm applications with Gemini CLI + +minutes_to_complete: 60 + +who_is_this_for: This is an advanced topic for developers and platform engineers who want to use Gemini CLI to analyze, migrate, and optimize applications for Arm-based systems. + +learning_objectives: + - Install and authenticate Gemini CLI on Arm Linux or macOS + - Configure Gemini CLI with persistent context for Arm development + - Integrate the Arm Model Context Protocol (MCP) server with Gemini CLI + - Use Gemini CLI to analyze code and workflows for Arm compatibility + - Apply Gemini CLI to common Arm migration and optimization scenarios + +prerequisites: + - Gemini CLI installed and authenticated + - Familiarity with Linux or macOS command-line tools + - General understanding of Arm architecture concepts + +author: Jason Andrews + +##### Tags +skilllevels: Advanced +subjects: Developer Tools +cloud_service_providers: Google Cloud + +tools_software_languages: + - Gemini CLI + - Docker + +operatingsystems: + - Linux + - macOS + +# ================================================================================ +# FIXED, DO NOT MODIFY +# ================================================================================ +further_reading: + - resource: + title: Gemini API documentation + link: https://ai.google.dev/gemini-api/docs + type: documentation + + - resource: + title: Gemini CLI documentation + link: https://ai.google.dev/gemini-api/docs/cli + type: documentation + + - resource: + title: Arm Developer documentation + link: https://developer.arm.com/documentation + type: documentation + + - resource: + title: Automate x86-to-Arm application migration using Arm MCP Server + link: https://learn.arm.com/learning-paths/servers-and-cloud-computing/arm-mcp-server/ + type: Install Guide +weight: 1 +layout: "learningpathall" +learning_path_main_page: "yes" +--- \ No newline at end of file diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/_next-steps.md b/content/learning-paths/servers-and-cloud-computing/gemini/_next-steps.md new file mode 100644 index 0000000000..c3db0de5a2 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/_next-steps.md @@ -0,0 +1,8 @@ +--- +# ================================================================================ +# FIXED, DO NOT MODIFY THIS FILE +# ================================================================================ +weight: 21 # Set to always be larger than the content in this path to be at the end of the navigation. +title: "Next Steps" # Always the same, html page title. +layout: "learningpathall" # All files under learning paths have this same wrapper for Hugo processing. +--- diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/arm-code-analysis.md b/content/learning-paths/servers-and-cloud-computing/gemini/arm-code-analysis.md new file mode 100644 index 0000000000..749453f012 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/arm-code-analysis.md @@ -0,0 +1,42 @@ + +--- +title: Use Gemini CLI for Arm code analysis and migration +weight: 7 + +### FIXED, DO NOT MODIFY +layout: learningpathall +--- + +## Overview + +In this section, you use Gemini CLI to reason about Arm compatibility and migration considerations for existing codebases. + +Gemini CLI provides guidance and analysis, not automated code transformation. + +## Analyze code portability + +Ask Gemini questions such as: + +```text +Are there common x86 assumptions in this codebase? +``` + +```text +What should I check when migrating this application to Arm64? +``` + +Gemini can help identify: +- Architecture-specific dependencies +- Build system assumptions +- Potential portability risks + +## Migration guidance + +Use Gemini CLI to reason about: +- Compiler flags +- Dependency availability +- Runtime behavior differences + +This guidance helps you plan migrations more effectively. + +Next, you explore how Gemini CLI can assist with optimization-related questions. diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/background.md b/content/learning-paths/servers-and-cloud-computing/gemini/background.md new file mode 100644 index 0000000000..b54e66b1c9 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/background.md @@ -0,0 +1,29 @@ +--- +title: Get started with Gemini CLI for Arm development + +weight: 2 + +layout: "learningpathall" +--- + +## The benefits of Arm-based development environments + +Arm-based systems are widely used across cloud, edge, and embedded environments due to their performance, efficiency, and scalability. Developers targeting Arm platforms often work across a mix of Linux and macOS systems and need tooling that understands Arm architecture constraints and best practices. + +Gemini CLI is designed to run consistently across operating systems, including Arm Linux distributions and macOS on Arm. This makes it a practical command-line assistant for developers building, migrating, or optimizing software for Arm-based environments. + +## Discover Gemini CLI + +Gemini CLI is Google’s command-line interface for interacting with the Gemini AI assistant. It allows developers to ask questions, explore design options, and reason about code and systems directly from the terminal. + +Gemini CLI supports tasks such as: +- Understanding unfamiliar codebases +- Reasoning about architecture and design decisions +- Exploring portability and migration considerations +- Assisting with development and troubleshooting workflows + +Because Gemini CLI is architecture-agnostic and runs on Arm-based systems, it can be used as part of an Arm-native development workflow without relying on x86-only tools. + +In this Learning Path, you configure Gemini CLI with Arm-specific context and integrate additional tooling so that Gemini can provide more relevant guidance for Arm development tasks. + +For more information, see the [Gemini CLI documentation](https://ai.google.dev/gemini-api/docs/cli). \ No newline at end of file diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/configure-context.md b/content/learning-paths/servers-and-cloud-computing/gemini/configure-context.md new file mode 100644 index 0000000000..551fe86522 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/configure-context.md @@ -0,0 +1,56 @@ + +--- +title: Configure Gemini CLI context for Arm development +weight: 4 + +### FIXED, DO NOT MODIFY +layout: learningpathall +--- + +## Overview + +In this section, you configure Gemini CLI with persistent context so it can provide more relevant guidance for Arm development tasks. + +By defining your development environment, architecture preferences, and constraints, you help Gemini generate responses that are tailored to Arm-based systems. + +## Create the Gemini configuration directory + +Gemini CLI reads context files from the `.gemini` directory in your home folder. + +Create the directory if it does not already exist: + +```console +mkdir -p ~/.gemini +``` + +## Define Arm development context + +Create a context file named `GEMINI.md` that describes your Arm development environment. + +```console +cat > ~/.gemini/GEMINI.md << 'EOF' +I am an Arm Linux developer. +I primarily work on Arm64 systems and prefer Arm-native solutions. +Please avoid x86-specific assumptions unless explicitly requested. +EOF +``` + +This file provides Gemini with persistent context that is automatically applied to all future sessions. + +## Verify that context is applied + +Start Gemini CLI: + +```console +gemini +``` + +Ask a question related to development tooling: + +```text +What should I consider when building software for Arm? +``` + +If context is loaded correctly, Gemini responds with Arm-specific guidance. + +Next, you learn how Gemini CLI can be extended using the Arm Model Context Protocol. diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/install.md b/content/learning-paths/servers-and-cloud-computing/gemini/install.md new file mode 100644 index 0000000000..36a8af5839 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/install.md @@ -0,0 +1,164 @@ +--- +title: Install Gemini CLI +weight: 3 + +### FIXED, DO NOT MODIFY +layout: learningpathall +--- + +## Overview + +In this section, you install Gemini CLI on macOS or Arm Linux and verify that it is working. + +Gemini CLI is Google’s command-line interface for interacting with the Gemini AI assistant from your terminal. + +## Before you begin + +To use Gemini CLI, you need: + +- A Google account +- An authentication method for Gemini (OAuth login is recommended for most users) +- Node.js 20 or later (Gemini CLI runs on Node.js) + +## Set up authentication + +Gemini CLI supports multiple authentication methods. Choose one option. + +### Option 1: Google OAuth login + +This is the simplest option for most users. + +After you install Gemini CLI, run: + +```console +gemini +``` + +When prompted, select **Login with Google** and complete authentication in your browser. + +### Option 2: Gemini API key + +If you prefer using an API key (for example, in scripts or CI), generate one in Google AI Studio: + +1. Go to https://aistudio.google.com/apikey +2. Sign in with your Google account +3. Select **Create API Key** +4. Copy the generated key + +Set the key as an environment variable: + +```console +export GEMINI_API_KEY="YOUR_API_KEY" +``` + +### Option 3: Vertex AI (enterprise) + +If you use Gemini through Vertex AI, set these environment variables: + +```console +export GOOGLE_API_KEY="YOUR_GOOGLE_CLOUD_API_KEY" +export GOOGLE_GENAI_USE_VERTEXAI=true +``` + +## Install Gemini CLI on macOS + +You can install Gemini CLI on macOS using Homebrew or npm. + +### Install using Homebrew + +If you do not have Homebrew installed, see https://brew.sh/. + +Install Gemini CLI: + +```console +brew install gemini-cli +``` + +### Install using npm + +If you prefer npm, install Gemini CLI globally. + +1. Install Node.js 20 or later: + +```console +brew install node +``` + +2. Verify your Node.js version: + +```console +node --version +``` + +3. Install Gemini CLI: + +```console +npm install -g @google/gemini-cli +``` + +## Install Gemini CLI on Arm Linux + +This section uses Ubuntu or Debian as the example. For other distributions, use the equivalent package manager. + +## Install prerequisites + +Install `curl`: + +```console +sudo apt update +sudo apt install -y curl +``` + +### Install Node.js 20 or later + +Use the NodeSource setup script: + +```console +curl -fsSL https://deb.nodesource.com/setup_20.x | sudo -E bash - +sudo apt install -y nodejs +``` + +Verify Node.js and npm: + +```console +node --version +npm --version +``` + +### Install Gemini CLI + +Install Gemini CLI globally: + +```console +sudo npm install -g @google/gemini-cli +``` + +## Verify the installation + +Confirm that the CLI is available: + +```console +gemini --version +``` + +Start an interactive session: + +```console +gemini +``` + +On first run, you are prompted to authenticate if you have not already done so. + +## View available CLI options + +To print the available commands and options, use: + +```console +gemini --help +``` + +## What you have accomplished and what is next + +You have installed Gemini CLI and verified that it runs on your system. + +Next, you can configure Gemini CLI with persistent context for Arm development workflows, and optionally integrate additional tooling such as an MCP server. diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/integrate.md b/content/learning-paths/servers-and-cloud-computing/gemini/integrate.md new file mode 100644 index 0000000000..15f26ce152 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/integrate.md @@ -0,0 +1,77 @@ + +--- +title: Integrate the Arm MCP server with Gemini CLI +weight: 6 + +### FIXED, DO NOT MODIFY +layout: learningpathall +--- + +## Overview + +In this section, you integrate the Arm Model Context Protocol (MCP) server with Gemini CLI. + +This enables Gemini to access Arm-specific tools and documentation during interactive sessions. + +## Prerequisites + +Before continuing, ensure: +- Gemini CLI is installed and working +- Docker is installed and running on your system + +## Pull the Arm MCP server image + +Pull the latest Arm MCP server container image: + +```console +docker pull armlimited/arm-mcp:latest +``` + +## Configure Gemini CLI to use MCP + +Edit the Gemini CLI settings file: + +```console +mkdir -p ~/.gemini +nano ~/.gemini/settings.json +``` + +Add the following configuration: + +```json +{ + "mcpServers": { + "arm_mcp_server": { + "command": "docker", + "args": [ + "run", + "--rm", + "-i", + "-v", "/path/to/your/code:/workspace", + "armlimited/arm-mcp:latest" + ], + "timeout": 60000 + } + } +} +``` + +Replace `/path/to/your/code` with the path to your local source directory. + +## Verify MCP integration + +Start Gemini CLI: + +```console +gemini +``` + +List available tools: + +```text +/tools +``` + +If the Arm MCP server is configured correctly, Arm-related tools appear in the output. + +Next, you use Gemini CLI with MCP to analyze Arm code and migration scenarios. diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/optimization.md b/content/learning-paths/servers-and-cloud-computing/gemini/optimization.md new file mode 100644 index 0000000000..9f02b24e71 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/optimization.md @@ -0,0 +1,42 @@ + +--- +title: Use Gemini CLI for Arm optimization guidance +weight: 8 + +### FIXED, DO NOT MODIFY +layout: learningpathall +--- + +## Overview + +In this section, you explore how Gemini CLI can assist with Arm optimization discussions and decision-making. + +Gemini CLI complements traditional tools such as profilers and benchmarks. + +## Ask optimization-focused questions + +Examples include: + +```text +What performance considerations are important on Arm? +``` + +```text +Where should I start when optimizing this workload for Arm? +``` + +Gemini can suggest: +- Areas to investigate +- Relevant Arm architecture concepts +- Appropriate tooling + +## Set expectations + +Gemini CLI does not replace: +- Performance profilers +- Benchmarks +- Hardware-specific tuning + +Instead, it helps you ask better questions and focus your optimization efforts. + +You have now completed the Gemini CLI Learning Path. diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md b/content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md new file mode 100644 index 0000000000..3dd402d413 --- /dev/null +++ b/content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md @@ -0,0 +1,38 @@ + +--- +title: Understand the Arm Model Context Protocol +weight: 5 + +### FIXED, DO NOT MODIFY +layout: learningpathall +--- + +## Overview + +In this section, you learn what the Arm Model Context Protocol (MCP) is and how it extends Gemini CLI with Arm-specific tools and knowledge. + +This page focuses on concepts only. You do not configure or run MCP yet. + +## What is the Model Context Protocol (MCP)? + +The Model Context Protocol allows Gemini CLI to interact with external tools that provide structured data, documentation, and analysis capabilities. + +Instead of relying only on a language model, Gemini can call MCP tools to retrieve authoritative information or perform targeted analysis. + +## Why use MCP for Arm development? + +The Arm MCP server provides: +- Access to Arm-specific documentation +- Architecture-aware guidance +- Tools for reasoning about migration and optimization + +MCP is most useful when you want deeper, architecture-aware assistance beyond general development advice. + +## When MCP is not required + +You do not need MCP for: +- General programming questions +- Language syntax help +- Non-architecture-specific tasks + +In the next section, you configure Gemini CLI to use the Arm MCP server. From 9aa573453334de17ecb970d6fb2f7f7da41b5569 Mon Sep 17 00:00:00 2001 From: Madeline Underwood Date: Tue, 30 Dec 2025 19:32:53 +0000 Subject: [PATCH 2/5] Refactor learning path content for Gemini CLI: update titles and enhance clarity in installation and MCP sections --- .../gemini/_index.md | 2 +- .../gemini/background.md | 10 +- .../gemini/install.md | 152 +++--------------- .../gemini/understand-mcp.md | 20 ++- 4 files changed, 42 insertions(+), 142 deletions(-) diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md index 3ff6496c73..120e3f4df6 100644 --- a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md +++ b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md @@ -6,7 +6,7 @@ minutes_to_complete: 60 who_is_this_for: This is an advanced topic for developers and platform engineers who want to use Gemini CLI to analyze, migrate, and optimize applications for Arm-based systems. learning_objectives: - - Install and authenticate Gemini CLI on Arm Linux or macOS + - Verify an existing Gemini CLI installation and authentication - Configure Gemini CLI with persistent context for Arm development - Integrate the Arm Model Context Protocol (MCP) server with Gemini CLI - Use Gemini CLI to analyze code and workflows for Arm compatibility diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/background.md b/content/learning-paths/servers-and-cloud-computing/gemini/background.md index b54e66b1c9..a870cbb6f0 100644 --- a/content/learning-paths/servers-and-cloud-computing/gemini/background.md +++ b/content/learning-paths/servers-and-cloud-computing/gemini/background.md @@ -6,24 +6,24 @@ weight: 2 layout: "learningpathall" --- -## The benefits of Arm-based development environments +## Developing software for Arm platforms Arm-based systems are widely used across cloud, edge, and embedded environments due to their performance, efficiency, and scalability. Developers targeting Arm platforms often work across a mix of Linux and macOS systems and need tooling that understands Arm architecture constraints and best practices. -Gemini CLI is designed to run consistently across operating systems, including Arm Linux distributions and macOS on Arm. This makes it a practical command-line assistant for developers building, migrating, or optimizing software for Arm-based environments. +As Arm adoption grows, developers often need to reason about portability, build system assumptions, dependency availability, and performance characteristics that differ from traditional x86 environments. These considerations are especially relevant when migrating existing applications or optimizing workloads to run efficiently on Arm-based systems. -## Discover Gemini CLI +## Gemini CLI Gemini CLI is Google’s command-line interface for interacting with the Gemini AI assistant. It allows developers to ask questions, explore design options, and reason about code and systems directly from the terminal. +Gemini CLI is designed to run consistently across operating systems, including Arm Linux distributions and macOS on Arm. This makes it a practical command-line assistant for developers building, migrating, or optimizing software for Arm-based environments. + Gemini CLI supports tasks such as: - Understanding unfamiliar codebases - Reasoning about architecture and design decisions - Exploring portability and migration considerations - Assisting with development and troubleshooting workflows -Because Gemini CLI is architecture-agnostic and runs on Arm-based systems, it can be used as part of an Arm-native development workflow without relying on x86-only tools. - In this Learning Path, you configure Gemini CLI with Arm-specific context and integrate additional tooling so that Gemini can provide more relevant guidance for Arm development tasks. For more information, see the [Gemini CLI documentation](https://ai.google.dev/gemini-api/docs/cli). \ No newline at end of file diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/install.md b/content/learning-paths/servers-and-cloud-computing/gemini/install.md index 36a8af5839..30701db498 100644 --- a/content/learning-paths/servers-and-cloud-computing/gemini/install.md +++ b/content/learning-paths/servers-and-cloud-computing/gemini/install.md @@ -1,5 +1,5 @@ --- -title: Install Gemini CLI +title: Verify and prepare Gemini CLI weight: 3 ### FIXED, DO NOT MODIFY @@ -8,157 +8,49 @@ layout: learningpathall ## Overview -In this section, you install Gemini CLI on macOS or Arm Linux and verify that it is working. +In this section, you verify that Gemini CLI is installed and authenticated, and confirm that it is ready for use in this Learning Path. -Gemini CLI is Google’s command-line interface for interacting with the Gemini AI assistant from your terminal. +{{% notice Note %}}This Learning Path assumes that Gemini CLI has already been installed. Installation steps are provided separately in the [Gemini CLI install guide](/install-guides/gemini/). +{{% /notice %}} ## Before you begin -To use Gemini CLI, you need: +To continue with this Learning Path, ensure that: -- A Google account -- An authentication method for Gemini (OAuth login is recommended for most users) -- Node.js 20 or later (Gemini CLI runs on Node.js) +- Gemini CLI is installed on your system +- You have completed authentication (for example, using Google OAuth or an API key) +- You can access a terminal on macOS or Arm Linux -## Set up authentication +## Verify Gemini CLI installation -Gemini CLI supports multiple authentication methods. Choose one option. - -### Option 1: Google OAuth login - -This is the simplest option for most users. - -After you install Gemini CLI, run: - -```console -gemini -``` - -When prompted, select **Login with Google** and complete authentication in your browser. - -### Option 2: Gemini API key - -If you prefer using an API key (for example, in scripts or CI), generate one in Google AI Studio: - -1. Go to https://aistudio.google.com/apikey -2. Sign in with your Google account -3. Select **Create API Key** -4. Copy the generated key - -Set the key as an environment variable: - -```console -export GEMINI_API_KEY="YOUR_API_KEY" -``` - -### Option 3: Vertex AI (enterprise) - -If you use Gemini through Vertex AI, set these environment variables: - -```console -export GOOGLE_API_KEY="YOUR_GOOGLE_CLOUD_API_KEY" -export GOOGLE_GENAI_USE_VERTEXAI=true -``` - -## Install Gemini CLI on macOS - -You can install Gemini CLI on macOS using Homebrew or npm. - -### Install using Homebrew - -If you do not have Homebrew installed, see https://brew.sh/. - -Install Gemini CLI: - -```console -brew install gemini-cli -``` - -### Install using npm - -If you prefer npm, install Gemini CLI globally. - -1. Install Node.js 20 or later: - -```console -brew install node -``` - -2. Verify your Node.js version: - -```console -node --version -``` - -3. Install Gemini CLI: - -```console -npm install -g @google/gemini-cli -``` - -## Install Gemini CLI on Arm Linux - -This section uses Ubuntu or Debian as the example. For other distributions, use the equivalent package manager. - -## Install prerequisites - -Install `curl`: - -```console -sudo apt update -sudo apt install -y curl -``` - -### Install Node.js 20 or later - -Use the NodeSource setup script: +Confirm that the Gemini CLI binary is available by checking the version: ```console -curl -fsSL https://deb.nodesource.com/setup_20.x | sudo -E bash - -sudo apt install -y nodejs +gemini --version ``` -Verify Node.js and npm: +Output similar to the following indicates that Gemini CLI is installed correctly: -```console -node --version -npm --version +```output +0.20.0 ``` -### Install Gemini CLI +## Verify authentication and start a session -Install Gemini CLI globally: - -```console -sudo npm install -g @google/gemini-cli -``` - -## Verify the installation - -Confirm that the CLI is available: - -```console -gemini --version -``` - -Start an interactive session: +Start an interactive Gemini CLI session: ```console gemini ``` -On first run, you are prompted to authenticate if you have not already done so. - -## View available CLI options +On first run, you are prompted to authenticate if authentication has not already been configured. After successful authentication, the interactive Gemini CLI session opens. -To print the available commands and options, use: - -```console -gemini --help -``` +If authentication fails or you need to change your authentication method, follow the instructions in the +[Gemini CLI Install Guide](/install-guides/gemini-cli/). ## What you have accomplished and what is next -You have installed Gemini CLI and verified that it runs on your system. +You have verified that Gemini CLI is installed and ready to use. + +Next, you configure Gemini CLI with persistent context so it can provide more relevant guidance for Arm development tasks. -Next, you can configure Gemini CLI with persistent context for Arm development workflows, and optionally integrate additional tooling such as an MCP server. diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md b/content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md index 3dd402d413..60ffe104e8 100644 --- a/content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md +++ b/content/learning-paths/servers-and-cloud-computing/gemini/understand-mcp.md @@ -1,6 +1,6 @@ --- -title: Understand the Arm Model Context Protocol +title: Explore the Arm Model Context Protocol weight: 5 ### FIXED, DO NOT MODIFY @@ -11,20 +11,26 @@ layout: learningpathall In this section, you learn what the Arm Model Context Protocol (MCP) is and how it extends Gemini CLI with Arm-specific tools and knowledge. +{{% notice Note %}} This page focuses on concepts only. You do not configure or run MCP yet. +{{% /notice %}} ## What is the Model Context Protocol (MCP)? The Model Context Protocol allows Gemini CLI to interact with external tools that provide structured data, documentation, and analysis capabilities. -Instead of relying only on a language model, Gemini can call MCP tools to retrieve authoritative information or perform targeted analysis. +Instead of relying only on a language model, Gemini CLI can call MCP tools to retrieve authoritative information or perform targeted analysis. + +{{% notice Tip %}} +You can think of MCP as a bridge between Gemini CLI and external, authoritative tools. Instead of relying solely on generated responses, Gemini CLI can use MCP to query structured sources and tools that are specifically designed for Arm development. +{{% /notice %}} ## Why use MCP for Arm development? -The Arm MCP server provides: -- Access to Arm-specific documentation -- Architecture-aware guidance -- Tools for reasoning about migration and optimization +When MCP is enabled, Gemini CLI can: +- Reference Arm-specific documentation and guidance more reliably +- Provide architecture-aware reasoning instead of generic advice +- Use specialized tools to support migration and optimization discussions MCP is most useful when you want deeper, architecture-aware assistance beyond general development advice. @@ -35,4 +41,6 @@ You do not need MCP for: - Language syntax help - Non-architecture-specific tasks +MCP does not replace compilers, profilers, or benchmarking tools. It augments Gemini CLI by providing better context and tooling for reasoning about Arm-specific development challenges. + In the next section, you configure Gemini CLI to use the Arm MCP server. From 38b789e55641f2231463d69e3761679e1244a6e4 Mon Sep 17 00:00:00 2001 From: Madeline Underwood Date: Tue, 30 Dec 2025 20:50:18 +0000 Subject: [PATCH 3/5] Added taxonomy fields for Gemini LP --- .../servers-and-cloud-computing/gemini/_index.md | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md index 120e3f4df6..da666f5670 100644 --- a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md +++ b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md @@ -21,16 +21,15 @@ author: Jason Andrews ##### Tags skilllevels: Advanced -subjects: Developer Tools -cloud_service_providers: Google Cloud - +subjects: Performance and Architecture +armips: + - Neoverse tools_software_languages: - - Gemini CLI - - Docker - + - Gemini CLI + - Docker operatingsystems: - - Linux - - macOS + - Linux + - macOS # ================================================================================ # FIXED, DO NOT MODIFY From e72dd3a77f9a0f2f1d0fc9e6c40d745c10a75780 Mon Sep 17 00:00:00 2001 From: Madeline Underwood Date: Tue, 30 Dec 2025 21:01:28 +0000 Subject: [PATCH 4/5] Fix formatting and update resource type in Gemini CLI learning path --- .../servers-and-cloud-computing/gemini/_index.md | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md index da666f5670..affd8e9171 100644 --- a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md +++ b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md @@ -1,5 +1,5 @@ --- -title: Analyze and optimize Arm applications with Gemini CLI +title: Analyze and optimize Arm applications with Gemini CLI minutes_to_complete: 60 @@ -22,14 +22,12 @@ author: Jason Andrews ##### Tags skilllevels: Advanced subjects: Performance and Architecture -armips: - - Neoverse tools_software_languages: - - Gemini CLI - - Docker + - Gemini CLI + - Docker operatingsystems: - - Linux - - macOS + - Linux + - macOS # ================================================================================ # FIXED, DO NOT MODIFY @@ -53,7 +51,7 @@ further_reading: - resource: title: Automate x86-to-Arm application migration using Arm MCP Server link: https://learn.arm.com/learning-paths/servers-and-cloud-computing/arm-mcp-server/ - type: Install Guide + type: documentation weight: 1 layout: "learningpathall" learning_path_main_page: "yes" From 079d090cdc613256238ed44476501834a4ee6397 Mon Sep 17 00:00:00 2001 From: Madeline Underwood Date: Tue, 30 Dec 2025 21:07:16 +0000 Subject: [PATCH 5/5] Add Arm IPs section to Gemini CLI learning path --- .../learning-paths/servers-and-cloud-computing/gemini/_index.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md index affd8e9171..2662f83cb9 100644 --- a/content/learning-paths/servers-and-cloud-computing/gemini/_index.md +++ b/content/learning-paths/servers-and-cloud-computing/gemini/_index.md @@ -25,6 +25,8 @@ subjects: Performance and Architecture tools_software_languages: - Gemini CLI - Docker +armips: + - Neoverse operatingsystems: - Linux - macOS