Require AI agents to attribute their GitHub comments (#1567)

* Attribute Claude-authored GitHub comments

gh posts comments as the authenticated user, making AI-generated
review/issue comments indistinguishable from the human's own. Requiring
a robot prefix keeps authorship transparent to reviewers.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* Make AI comment attribution agent-agnostic

Per review, the Copilot-facing instruction file should not hard-code
"Claude" attribution. A generic prefix keeps both files correct for
whichever agent authors the comment.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* Generalize comment-attribution rule to any authenticated user

Drops the hard-coded username so the guidance reads correctly for any
contributor whose gh account posts the comment.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* Tell agents to keep GitHub comments brief

Verbose AI comments add noise to PR threads; a brevity rule keeps them
scannable and pushes longer context into the PR description.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

Author: maebeale

.github/copilot-instructions.md

@@ -180,6 +180,8 @@ Follow the [Stimulus Handbook](https://stimulus.hotwired.dev/handbook/introducti
 - **On every push**, update the PR title and content to reflect the current diff — preserve any existing images/screenshots in the description
 - **On every push**, update AI instruction files if the diff adds, removes, or renames anything tracked in AGENTS.md — specifically: Stimulus controllers, services, model/controller concerns, mailers, rake tasks, and directory file counts
 - **On every push**, add PR review comments on notable lines of code — decisions, trade-offs, non-obvious logic, or anything a reviewer should understand. Use `gh api` to post line comments on the diff
+- **Attribute every AI-authored GitHub comment** — `gh` posts as the authenticated user, so any comment you create (PR review comments, issue/PR comments, replies) MUST be prefixed to identify the AI agent that wrote it. Begin the comment body with `🤖 _From <agent>:_` (e.g. `🤖 _From Claude:_` or `🤖 _From Copilot:_`) followed by the content
+- **Keep GitHub comments short and to the point** — one or two sentences, stating the key insight directly. Skip preamble, restating the code, and hedging; if a comment needs more than a few lines, it usually belongs in the PR description instead
 
 ## Testing
 

CLAUDE.md

@@ -180,6 +180,8 @@ Follow the [Stimulus Handbook](https://stimulus.hotwired.dev/handbook/introducti
 - **On every push**, update the PR title and content to reflect the current diff — preserve any existing images/screenshots in the description
 - **On every push**, update AI instruction files if the diff adds, removes, or renames anything tracked in AGENTS.md — specifically: Stimulus controllers, services, model/controller concerns, mailers, rake tasks, and directory file counts
 - **On every push**, add PR review comments on notable lines of code — decisions, trade-offs, non-obvious logic, or anything a reviewer should understand. Use `gh api` to post line comments on the diff
+- **Attribute every AI-authored GitHub comment** — `gh` posts as the authenticated user, so any comment you create (PR review comments, issue/PR comments, replies) MUST be prefixed to identify the AI agent that wrote it. Begin the comment body with `🤖 _From <agent>:_` (e.g. `🤖 _From Claude:_` or `🤖 _From Copilot:_`) followed by the content
+- **Keep GitHub comments short and to the point** — one or two sentences, stating the key insight directly. Skip preamble, restating the code, and hedging; if a comment needs more than a few lines, it usually belongs in the PR description instead
 
 ## Ruby Version Manager