Update prompt file + custom agents examples in customization library (#59070)

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

Author: saritai

content/copilot/tutorials/customization-library/custom-agents/bug-fix-teammate.md

@@ -61,4 +61,19 @@ You are a bug-fixing specialist focused on resolving issues in the codebase with
 Your goal is to make the codebase more stable and reliable by implementing working fixes, not just identifying problems.
 ```
 
+## How to use this {% data variables.copilot.copilot_custom_agent_short %}
+
+1. Go to the agents tab at [https://github.com/copilot/agents](https://github.com/copilot/agents?ref_product=copilot&ref_type=engagement&ref_style=text).
+1. Using the dropdown menus in the text box, select the repository and branch you want the custom agent to work in.
+1. Click {% octicon "copilot" aria-hidden="true" aria-label="copilot" %}, then click **{% octicon "plus" aria-label="Plus button" %} Create a custom agent**.
+1. An {% data variables.copilot.agent_profile %} template called `my-agent.agent.md` will open in the `.github/agents` directory, in the repository you chose. Name the file `bug-fix-teammate.agent.md` and paste in the example {% data variables.copilot.agent_profile %}.
+1. Commit and merge this file into your repository's default branch. Go back to the agents tab (you may need to refresh the page), and in the text box, select your "bug-fix-teammate" agent from the dropdown.
+1. In the text box, enter a task for the agent (such as the example below) and click **{% octicon "paper-airplane" aria-label="Start task" %} Start task** or press <kbd>Return</kbd>.
+
+   ```copilot copy
+   Scan the repository for the most critical bug, then implement a targeted fix and explain your approach.
+   ```
+
+The agent task will appear on the page below the text box. You can click into the task and follow along with the agent. For more information, see [AUTOTITLE](/copilot/how-tos/use-copilot-agents/coding-agent/track-copilot-sessions).
+
 {% data reusables.copilot.custom-agents-further-reading %}

content/copilot/tutorials/customization-library/custom-agents/cleanup-specialist.md

@@ -71,4 +71,19 @@ You are a cleanup specialist focused on making codebases cleaner and more mainta
 Focus on cleaning up existing code rather than adding new features. Work on both code files (.js, .py, etc.) and documentation files (.md, .txt, etc.) when removing duplication and improving consistency.
 ```
 
+## How to use this {% data variables.copilot.copilot_custom_agent_short %}
+
+1. Go to the agents tab at [https://github.com/copilot/agents](https://github.com/copilot/agents?ref_product=copilot&ref_type=engagement&ref_style=text).
+1. Using the dropdown menus in the text box, select the repository and branch you want the custom agent to work in.
+1. Click {% octicon "copilot" aria-hidden="true" aria-label="copilot" %}, then click **{% octicon "plus" aria-label="Plus button" %} Create a custom agent**.
+1. An {% data variables.copilot.agent_profile %} template called `my-agent.agent.md` will open in the `.github/agents` directory, in the repository you chose. Name the file `cleanup-specialist.agent.md` and paste in the example {% data variables.copilot.agent_profile %}.
+1. Commit and merge this file into your repository's default branch. Go back to the agents tab (you may need to refresh the page), and in the text box, select your "cleanup-specialist" agent from the dropdown.
+1. In the text box, enter a task for the agent (such as the example below) and click **{% octicon "paper-airplane" aria-label="Start task" %} Start task** or press <kbd>Return</kbd>.
+
+   ```copilot copy
+    Refactor and clean up any messy or inconsistent code in the `src/utils` directory, then document the improvements you made.
+   ```
+
+The agent task will appear on the page below the text box. You can click into the task and follow along with the agent. For more information, see [AUTOTITLE](/copilot/how-tos/use-copilot-agents/coding-agent/track-copilot-sessions).
+
 {% data reusables.copilot.custom-agents-further-reading %}

content/copilot/tutorials/customization-library/custom-agents/implementation-planner.md

@@ -75,4 +75,19 @@ For each phase, list specific tasks with complexity estimates (Small/Medium/Larg
 Adjust the detail level based on your needs - solo projects might need less formal documentation, while team projects benefit from more thorough planning. Focus on creating a roadmap that helps you stay organized and make progress.
 ```
 
+## How to use this {% data variables.copilot.copilot_custom_agent_short %}
+
+1. Go to the agents tab at [https://github.com/copilot/agents](https://github.com/copilot/agents?ref_product=copilot&ref_type=engagement&ref_style=text).
+1. Using the dropdown menus in the text box, select the repository and branch you want the custom agent to work in.
+1. Click {% octicon "copilot" aria-hidden="true" aria-label="copilot" %}, then click **{% octicon "plus" aria-label="Plus button" %} Create a custom agent**.
+1. An {% data variables.copilot.agent_profile %} template called `my-agent.agent.md` will open in the `.github/agents` directory, in the repository you chose. Name the file `implementation-planner.agent.md` and paste in the example {% data variables.copilot.agent_profile %}.
+1. Commit and merge this file into your repository's default branch. Go back to the agents tab (you may need to refresh the page), and in the text box, select your "implementation-planner" agent from the dropdown.
+1. In the text box, enter a task for the agent (such as the example below) and click **{% octicon "paper-airplane" aria-label="Start task" %} Start task** or press <kbd>Return</kbd>.
+
+   ```copilot copy
+    Create a detailed implementation plan for adding user authentication to our web app, including technical approach, phases, and risk assessment.
+   ```
+
+The agent task will appear on the page below the text box. You can click into the task and follow along with the agent. For more information, see [AUTOTITLE](/copilot/how-tos/use-copilot-agents/coding-agent/track-copilot-sessions).
+
 {% data reusables.copilot.custom-agents-further-reading %}

content/copilot/tutorials/customization-library/custom-agents/your-first-custom-agent.md

@@ -74,7 +74,7 @@ Test this agent by giving it a task to complete:
 
 1. Go to the agents tab at [https://github.com/copilot/agents](https://github.com/copilot/agents?ref_product=copilot&ref_type=engagement&ref_style=text).
 1. Using the dropdown menus in the text box, select the repository and branch you're comfortable testing with (ideally one with a minimal or outdated README).
-1. Click {% octicon "copilot" aria-hidden="true" aria-label="copilot" %}, then click **{% octicon "plus" aria-label="Plus button" %} Create an agent**.
+1. Click {% octicon "copilot" aria-hidden="true" aria-label="copilot" %}, then click **{% octicon "plus" aria-label="Plus button" %} Create a custom agent**.
 1. An {% data variables.copilot.agent_profile %} template called `my-agent.agent.md` will open in the `.github/agents` directory, in the repository you chose. Name the file `readme-specialist.agent.md` and paste in the example {% data variables.copilot.agent_profile %}.
 1. Commit and merge this file into your repository's default branch. Go back to the agents tab (you may need to refresh the page), and in the text box, select your "readme-specialist" agent from the dropdown.
 1. In the text box, enter a task for the agent (such as the example below) and click **{% octicon "paper-airplane" aria-label="Start task" %} Start task** or press <kbd>Return</kbd>.

content/copilot/tutorials/customization-library/prompt-files/document-api.md

@@ -23,7 +23,7 @@ This prompt file generates OpenAPI 3.0 specifications for REST API endpoints by
 
 ```text copy
 ---
-mode: 'agent'
+agent: 'agent'
 description: 'Generate OpenAPI 3.0 specification for API endpoints'
 ---
 

content/copilot/tutorials/customization-library/prompt-files/generate-unit-tests.md

@@ -23,7 +23,7 @@ This prompt file generates focused unit tests for specific functions or methods,
 
 ```text copy
 ---
-mode: 'agent'
+agent: 'agent'
 description: 'Generate unit tests for selected functions or methods'
 ---
 

content/copilot/tutorials/customization-library/prompt-files/onboarding-plan.md

@@ -21,7 +21,7 @@ contentType: tutorials
 
 ```text copy
 ---
-mode: 'agent'
+agent: 'agent'
 description: 'Help new team members onboard with a phased plan and suggestions for first tasks.'
 ---
 

content/copilot/tutorials/customization-library/prompt-files/review-code.md

@@ -25,7 +25,7 @@ You can also use {% data variables.copilot.copilot_code-review_short %} in {% da
 
 ```text copy
 ---
-mode: 'agent'
+agent: 'agent'
 description: 'Perform a comprehensive code review'
 ---
 

content/copilot/tutorials/customization-library/prompt-files/your-first-prompt-file.md

@@ -32,7 +32,7 @@ Start with this simple prompt file that helps you write clear, well-documented c
 
 ```text copy
 ---
-mode: 'agent'
+agent: 'agent'
 description: 'Generate a clear code explanation with examples'
 ---