Convert embedded custom agents to prompt file format

.github/agents/add-safe-output-type.md → .github/prompts/add-safe-output-type.prompt.md

@@ -1,4 +1,5 @@
 ---
+mode: agent
 name: add-safe-output-type
 description: Adding a New Safe Output Type to GitHub Agentic Workflows
 ---

.github/agents/cli-consistency-checker.md → .github/prompts/cli-consistency-checker.prompt.md

@@ -1,4 +1,5 @@
 ---
+mode: agent
 name: cli-consistency-checker
 description: Agent specialized in inspecting the gh-aw CLI tool by running actual commands with --help flags to verify consistency, documentation accuracy, and user-facing help text.
 ---

pkg/cli/templates/create-agentic-workflow.md → .github/prompts/create-agentic-workflow.prompt.md

@@ -1,5 +1,5 @@
 ---
-name: create-agentic-workflow
+mode: agent
 description: Design agentic workflows using GitHub Agentic Workflows (gh-aw) extension with interactive guidance on triggers, tools, and security best practices.
 ---
 

pkg/cli/templates/create-shared-agentic-workflow.md → .github/prompts/create-shared-agentic-workflow.prompt.md

@@ -1,4 +1,5 @@
 ---
+mode: agent
 name: create-shared-agentic-workflow
 description: Create shared agentic workflow components that wrap MCP servers using GitHub Agentic Workflows (gh-aw) with Docker best practices.
 ---

.github/agents/debug-agentic-workflow.md → .github/prompts/debug-agentic-workflow.prompt.md

@@ -1,5 +1,5 @@
 ---
-name: debug-agentic-workflow
+mode: agent
 description: Debug and refine agentic workflows using gh-aw CLI tools - analyze logs, audit runs, and improve workflow performance
 ---
 

.github/agents/improve-json-schema-descriptions.md → .github/prompts/improve-json-schema-descriptions.prompt.md

@@ -1,4 +1,5 @@
 ---
+mode: agent
 name: improve-json-schema-descriptions
 description: Systematic approach for reviewing and improving descriptions in the frontmatter JSON schema for GitHub Agentic Workflows
 ---

pkg/cli/templates/setup-agentic-workflows.md → .github/prompts/setup-agentic-workflows.prompt.md

@@ -1,5 +1,5 @@
 ---
-name: setup-agentic-workflows
+mode: agent
 description: A guided agent to help you set up your agentic workflows using gh-aw
 ---
 
@@ -89,15 +89,11 @@ gh secret set OPENAI_API_KEY -a actions --body "your-openai-api-key-here"
 
 Say to the user:
 ````
-When you're ready, use the custom agent to create your workflow:
+When you're ready, use the prompt to create your workflow:
 
-**Option 1: Direct invocation**
+**Direct invocation**
 Type `/create-agentic-workflow` in the chat
 
-**Option 2: Menu selection**
-1. Type `/agent` in the chat
-2. Select `create-agentic-workflow` from the list of available custom agents
-
-This will activate the workflow creation custom agent to help you create your first agentic workflow.
+This will help you create your first agentic workflow.
 
 ````

.github/agents/shell-2-script.md → .github/prompts/shell-2-script.prompt.md

@@ -1,4 +1,5 @@
 ---
+mode: agent
 name: shell-2-script
 description: Extract inline bash scripts from Go compiler code into separate .sh files with embedded resources for improved maintainability, organization, and reusability
 ---

.github/agents/technical-doc-writer.md → .github/prompts/technical-doc-writer.prompt.md

@@ -1,4 +1,5 @@
 ---
+mode: agent
 name: technical-doc-writer
 description: AI technical documentation writer for GitHub Actions library using Astro Starlight and GitHub Docs voice
 ---

Makefile

@@ -286,9 +286,10 @@ sync-templates:
 	@echo "Syncing templates from .github to pkg/cli/templates..."
 	@mkdir -p pkg/cli/templates
 	@cp .github/aw/github-agentic-workflows.md pkg/cli/templates/
-	@cp .github/agents/create-agentic-workflow.md pkg/cli/templates/
-	@cp .github/agents/setup-agentic-workflows.md pkg/cli/templates/
-	@cp .github/agents/create-shared-agentic-workflow.md pkg/cli/templates/
+	@cp .github/prompts/create-agentic-workflow.prompt.md pkg/cli/templates/
+	@cp .github/prompts/setup-agentic-workflows.prompt.md pkg/cli/templates/
+	@cp .github/prompts/create-shared-agentic-workflow.prompt.md pkg/cli/templates/
+	@cp .github/prompts/debug-agentic-workflow.prompt.md pkg/cli/templates/
 	@echo "✓ Templates synced successfully"
 
 # Sync action pins from .github/aw to pkg/workflow/data

docs/src/content/docs/reference/custom-agents.md

@@ -49,6 +49,6 @@ The agent instructions are merged with the workflow prompt, customizing the AI e
 - **Frontmatter**: Can include `name`, `description`, `tools`, and `mcp-servers`
 - **One per workflow**: Only one agent file can be imported per workflow
 
-## Built-in Custom Agents
+## Built-in Prompts
 
-The `gh aw init` command sets up a `debug-agentic-workflow` custom agent that helps debug and refine workflows using CLI tools (`gh aw logs`, `gh aw audit`, `gh aw run`, `gh aw compile`). The agent provides conversational workflow debugging and performance analysis.
+The `gh aw init` command sets up workflow creation prompts including `debug-agentic-workflow` that helps debug and refine workflows using CLI tools (`gh aw logs`, `gh aw audit`, `gh aw run`, `gh aw compile`). These prompts are stored in `.github/prompts/` and provide conversational workflow authoring and debugging capabilities.

docs/src/content/docs/reference/glossary.md

@@ -100,6 +100,9 @@ A human-friendly data format used for configuration files. Uses indentation and
 ### Personal Access Token (PAT)
 A token that authenticates you to GitHub's APIs with specific permissions. Required for GitHub Copilot CLI to access Copilot services. Created at github.com/settings/personal-access-tokens.
 
+### Prompt Files
+Markdown files with YAML frontmatter stored in `.github/prompts/` that define interactive Copilot Chat commands. Created by `gh aw init`, these files (like `create-agentic-workflow.prompt.md`) can be invoked as slash commands (e.g., `/create-agentic-workflow`) in Copilot Chat to guide workflow creation with specialized instructions.
+
 ### Fine-grained Personal Access Token
 A type of GitHub Personal Access Token with granular permission control. Specify exactly which repositories the token can access and what permissions it has (`contents: read`, `issues: write`, etc.). Created at github.com/settings/personal-access-tokens.
 
@@ -123,7 +126,10 @@ Special triggers that respond to slash commands in issue and PR comments (e.g.,
 Settings that limit how many instances of a workflow can run simultaneously. Configured using the `concurrency:` field to prevent resource conflicts or rate limiting.
 
 ### Custom Agents
-Specialized instructions or configurations that customize AI agent behavior for specific tasks or repositories. Stored in `.github/agents/` or `.github/copilot/instructions/` directories.
+Specialized instructions or configurations that customize AI agent behavior for specific tasks or repositories. Can be stored as:
+- **Prompt files** (`.github/prompts/*.prompt.md`) - Used with Copilot Chat slash commands for interactive workflow authoring
+- **Agent files** (`.github/agents/*.md`) - Imported into workflows via the `imports` field for execution-time customization
+- **Instruction files** (`.github/copilot/instructions/`) - Path-specific or repository-wide Copilot instructions
 
 ### Strict Mode
 An enhanced validation mode that enforces additional security checks and best practices. Enabled using `strict: true` in frontmatter or the `--strict` flag when compiling.

docs/src/content/docs/setup/agentic-authoring.mdx

@@ -7,48 +7,48 @@ sidebar:
 
 import CopyEntireFileButton from '../../../components/CopyEntireFileButton.astro';
 
-GitHub Agentic Workflows provide a custom agent that turns GitHub Copilot into 
-a powerful workflow authoring tool. This guide explains how to use this agent to author agentic workflows in natural language.
+GitHub Agentic Workflows provide prompts that turn GitHub Copilot into 
+a powerful workflow authoring tool. This guide explains how to use these prompts to author agentic workflows in natural language.
 
 ## Quick Start
 
-Initialize your repository to set up GitHub Copilot instructions and custom agents for authoring agentic workflows:
+Initialize your repository to set up GitHub Copilot instructions and prompts for authoring agentic workflows:
 
 ```bash wrap
 gh aw init
 ```
 
 This creates:
 - `.github/aw/github-agentic-workflows.instructions.md` - Custom Copilot instructions for better workflow authoring
-- `.github/agents/create-agentic-workflow.md` - Custom agent for the `/agent` command in Copilot Chat
+- `.github/prompts/create-agentic-workflow.prompt.md` - Prompt for the `/create-agentic-workflow` command in Copilot Chat
 
 :::tip[VS Code integration]
 Once initialized, GitHub Copilot will automatically use these instructions when you edit workflow files in VS Code. See [VS Code Integration](/gh-aw/setup/vscode/) for more details.
 :::
 
-## What is the `create-agentic-workflow` custom agent?
+## What is the `create-agentic-workflow` prompt?
 
-`.github/agents/create-agentic-workflow.md` is a custom agent file that contains the structure and instructions GitHub Copilot will use to generate a workflow markdown file that the `gh aw` CLI understands. A [custom agent](/gh-aw/reference/glossary/#custom-agents) is a specialized instruction set that customizes AI behavior for specific tasks.
+`.github/prompts/create-agentic-workflow.prompt.md` is a prompt file that contains the structure and instructions GitHub Copilot will use to generate a workflow markdown file that the `gh aw` CLI understands. A [prompt file](/gh-aw/reference/glossary/#prompt-files) is a specialized instruction set that customizes AI behavior for specific tasks.
 
-Use the custom agent when you want to:
+Use the prompt when you want to:
 - Draft a new agentic workflow using natural language
 - Iterate on workflow steps with AI assistance inside your editor
 
-The custom agent contains specialized instructions and toolset to enable efficient workflow authoring.
+The prompt contains specialized instructions and toolset to enable efficient workflow authoring.
 
-## How to use the `create-agentic-workflow` custom agent
+## How to use the `create-agentic-workflow` prompt
 
-In Visual Studio Code with GitHub Copilot Chat or in Copilot CLI, enter the `/agent` command and select `create-agentic-workflow` from the list:
+In Visual Studio Code with GitHub Copilot Chat or in Copilot CLI, enter the `/create-agentic-workflow` command:
 
 ```sh wrap
-/agent
+/create-agentic-workflow
 ```
 
-Then select `create-agentic-workflow` from the available custom agents. This will start the agentic workflow authoring process with interactive guidance.
+This will start the agentic workflow authoring process with interactive guidance.
 
 ## Other Agents and Chats
 
-The custom agent file can be loaded into other AI chat interfaces that support custom instructions. The agent is designed to be compatible with various AI tools, although some features might require configuration and you'll need to allow running the compiler.
+The prompt file can be loaded into other AI chat interfaces that support custom instructions. The prompt is designed to be compatible with various AI tools, although some features might require configuration and you'll need to allow running the compiler.
 
 ## Dictating Agentic Workflows
 

docs/src/content/docs/setup/cli.md

@@ -78,11 +78,11 @@ Commands are organized by workflow lifecycle: creating, building, testing, monit
 Initialize your repository for agentic workflows.
 
 ```bash wrap
-gh aw init       # Configure .gitattributes, Copilot instructions, custom agent
+gh aw init       # Configure .gitattributes, Copilot instructions, prompts
 gh aw init --mcp # Also setup MCP server integration for Copilot Agent
 ```
 
-Configures `.gitattributes` to mark `.lock.yml` files as generated, adds Copilot instructions for better AI assistance, and sets up custom agent configuration. The `--mcp` flag additionally creates GitHub Actions workflow for MCP server setup, configures `.vscode/mcp.json` for VS Code integration, and enables gh-aw MCP tools in Copilot Agent.
+Configures `.gitattributes` to mark `.lock.yml` files as generated, adds Copilot instructions for better AI assistance, and sets up prompt files for workflow creation. The `--mcp` flag additionally creates GitHub Actions workflow for MCP server setup, configures `.vscode/mcp.json` for VS Code integration, and enables gh-aw MCP tools in Copilot Agent.
 
 #### `add`
 

docs/src/content/docs/setup/quick-start.md

@@ -166,7 +166,7 @@ You can leverage the help of an agent to customize your workflow without having
 ```bash wrap
 # install copilot cli
 npm install -g @github/copilot-cli
-# install the custom agent files
+# install the prompt files
 gh aw init
 ```
 
@@ -175,7 +175,7 @@ Then, run the following to create and edit your workflow:
 ```bash wrap
 # start an interactive session to customize the workflow
 copilot
-> /agent create-agentic-workflow
+> /create-agentic-workflow
 > edit @.github/workflows/daily-team-status.md
 ```
 

docs/src/content/docs/setup/vscode.md

@@ -17,11 +17,11 @@ gh aw init
 
 This creates files including:
 - **Copilot instructions**: `.github/aw/github-agentic-workflows.md` - Automatically imported when authoring markdown files under `.github/workflows/`, making Copilot Chat more efficient at generating Agentic Workflows.
-- **Custom agent**: `.github/agents/create-agentic-workflow.md` - A custom agent you can reference in Copilot Chat to interactively create workflows.
+- **Prompts**: `.github/prompts/create-agentic-workflow.prompt.md` - A prompt you can reference in Copilot Chat to interactively create workflows.
 
-## create-agentic-workflow custom agent <a id="create-agentic-workflow"></a>
+## create-agentic-workflow prompt <a id="create-agentic-workflow"></a>
 
-Use the `/agent` command in Copilot Chat and select `create-agentic-workflow` from the list to interactively design workflows with guided assistance for trigger selection, permissions, security settings, tool configuration, and best practices. The conversational interface helps you build secure workflows without memorizing syntax.
+Use the `/create-agentic-workflow` command in Copilot Chat to interactively design workflows with guided assistance for trigger selection, permissions, security settings, tool configuration, and best practices. The conversational interface helps you build secure workflows without memorizing syntax.
 
 ## Background Compilation