Skip to content

docs(aw): add "Creating Command Workflows" section to create-agentic-workflow.md#21410

Merged
pelikhan merged 4 commits intomainfrom
copilot/update-create-agent-workflows
Mar 17, 2026
Merged

docs(aw): add "Creating Command Workflows" section to create-agentic-workflow.md#21410
pelikhan merged 4 commits intomainfrom
copilot/update-create-agent-workflows

Conversation

Copy link
Contributor

Copilot AI commented Mar 17, 2026

The workflow creator agent lacked guidance on how to build on-demand "command" workflows, leaving it without a clear recommendation between the two supported approaches.

Changes

  • New ## Creating Command Workflows section in .github/aw/create-agentic-workflow.md covering both triggers with tradeoffs, when-to-recommend guidance, permission requirements, and working examples:

    • slash_command — user types /command in a comment; flexible, composable, accepts context via comment body, less discoverable
    • label_command — user applies a label; auto-removed after activation, visible in GitHub UI sidebar, no argument passing
  • Comparison table for at-a-glance decision making:

    slash_command label_command
    Invocation /command as first word of a comment Apply a label via GitHub UI
    Discoverability Low High — visible in label picker
    Arguments Comment body provides context No arguments; one-shot action
    Part of LabelOps No Yes
  • Combining both — documents the slash_command + label_command OR-trigger pattern so workflows can be reached via either mechanism

  • Cross-references updated — Issue Form Mode Step 2 trigger list gains an "On-demand commands" entry pointing to the new section; "Starting the conversation" clarifying question updated to mention label commands

Original prompt

Update the create-agent- workflows.md with a section on creating commands. Label_command and slash_command can be used with each their own tradeoff. Label is more precise and part of labelops. Slash_command is more flexible but less discoverable. These are the preferred ways to create "command" line workflows.


💬 Send tasks to Copilot coding agent from Slack and Teams to turn conversations into code. Copilot posts an update in your thread when it's finished.

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
Copilot AI changed the title [WIP] Update create-agent-workflows.md with command creation section docs(aw): add "Creating Command Workflows" section to create-agentic-workflow.md Mar 17, 2026
Copilot AI requested a review from pelikhan March 17, 2026 15:29
@pelikhan pelikhan marked this pull request as ready for review March 17, 2026 15:34
Copilot AI review requested due to automatic review settings March 17, 2026 15:34
@pelikhan pelikhan merged commit fb7d3e8 into main Mar 17, 2026
@pelikhan pelikhan deleted the copilot/update-create-agent-workflows branch March 17, 2026 15:34
Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Adds guidance for authoring on-demand “command” agentic workflows so workflow creators can choose the right trigger pattern and implement it correctly.

Changes:

  • Added a new “Creating Command Workflows” section documenting slash_command vs label_command, including tradeoffs, recommendations, and examples.
  • Added a comparison table and documented an OR-trigger pattern combining both triggers.
  • Updated existing guidance to reference label commands in the interactive prompt and Issue Form Mode trigger selection.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.

label_command: deploy
permissions:
contents: read
pull-requests: write # Required for automatic label removal
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants