Feature: built-in AI-authorship disclosure footer for safe-outputs (or shared snippet support)

[Evangelink]

Summary

Provide a first-class opt-in "AI authorship disclosure" footer that gh-aw automatically appends to every safe-output (issue body, PR body, discussion, comment, …). Today every workflow has to repeat the same safe-outputs.messages.footer: "…" boilerplate string.

Motivation

Across our 20+ agentic workflows we now repeat the exact same footer string on every single one:

safe-outputs:
  messages:
    footer: "> 🤖 **Automated content by GitHub Copilot.** Posted via a maintainer's GitHub token, so it appears under their account — the account owner did **not** write or approve this content personally. Generated by the [{workflow_name}]({agentic_workflow_url}) workflow.{ai_credits_suffix} · [◷]({history_link})"

Workaround PR: microsoft/testfx#8959 — "Add AI/Copilot disclaimer footer to all agentic workflows", which had to touch ~20 workflow files just to bolt on the same string.

The motivation for adding this footer is exactly the kind of transparency the platform should make easy:

• Safe-outputs run under a maintainer's PAT, so issues/PRs appear authored by that human even though the content is fully AI-generated.
• Without a disclosure footer, readers can't tell automated content from human content, which has real implications for code review trust and even legal/compliance disclosure requirements.

Proposed solution

Either of:

1. Built-in flag with sensible default text:

   safe-outputs:
     disclosure-footer: true     # appends a standard disclosure
     # or
     disclosure-footer:
       template: "..."           # override

2. Shared snippets / imports: for footer strings, so the boilerplate can live in one place and be imports:ed across workflows (this would also help other repeated frontmatter blocks).

Either way the goal is: one toggle on a single workflow, instead of N copies of the same string across N workflows.

Related

• Feature: Content redaction for public-facing outputs #34906 — Content redaction for public-facing outputs (adjacent: both about controlling what shows up in safe-outputs)

Environment

• gh-aw: v0.75.x

[pelikhan]

@Evangelink move the footer yaml into a share agentic workflow file and import it.

[Evangelink]

@Evangelink move the footer yaml into a share agentic workflow file and import it.

Right! Makes sense, I'll do on my side. I still think it'd be good to have it by default, as having PRs or comments done with user identity (user PAT) can lead to weird situations.

[pelikhan]

@copilot add a disclosure-header message that gets prepended to messages stating the content was ai generated.