[product-manager-impersonator] Threaded replies default for mention workflows

[github-actions]

Feature Idea

Summary: Add a small threading control input to mention-in-pr and mention-in-issue so replies can consistently target the triggering comment via add_comment.reply_to_id instead of always creating top-level comments.

Why a Customer Would Want This

Teams using /ai mention workflows want less timeline noise and clearer conversation context. Threaded replies keep the bot response attached to the exact request, which is easier to follow in busy PRs/issues and avoids comment pileups.

Rough Implementation Sketch

• Add a new optional workflow input (for example reply-in-thread) in the source workflows for:
  • .github/workflows/gh-aw-mention-in-pr.md
  • .github/workflows/gh-aw-mention-in-issue.md
• Update the prompt instructions so when enabled, the agent sets reply_to_id to the triggering comment ID on add_comment.
• Recompile lock workflows and update workflow READMEs to document the new behavior and default.
• Keep behavior backward-compatible (opt-in or carefully chosen default) so adopters can roll this out safely.

Why It Won't Be That Hard

This mostly wires existing capabilities together: safe outputs for both mention workflows already accept reply_to_id, so the work is confined to prompt/input plumbing and docs.

Evidence

• Mention workflow inputs currently have no threading control:
  • gh-agent-workflows/mention-in-pr/README.md#L26-L33
  • gh-agent-workflows/mention-in-issue/README.md#L25-L32
• Prompt instructions currently just tell the agent to call add_comment:
  • .github/workflows/gh-aw-mention-in-pr.md#L208-L211
  • .github/workflows/gh-aw-mention-in-issue.md#L133-L136
• Safe outputs already support threaded replies (reply_to_id) in these workflows:
  • .github/workflows/gh-aw-mention-in-pr.lock.yml#L844-L845 and #L867-L870
  • .github/workflows/gh-aw-mention-in-issue.lock.yml#L706-L707 and #L727-L730
• Recent product signal that comment-noise reduction matters:
  • Issue: PR Buildkite Detective creates one github comment per build failure on a PR #1098
  • PR: Reduce PR Buildkite Detective comment spam with marker-based upsert #1099

Note

🔒 Integrity filter blocked 47 items

The following items were blocked because they don't meet the GitHub integrity level.

• #937 search_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #1067 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #359 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #704 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #694 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #684 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #875 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #676 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #321 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #132 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #139 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #225 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #336 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #1025 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #354 search_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• Mint Ephemeral Tokens #1067 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• ... and 31 more items

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

---

What is this? | From workflow: Trigger Product Manager Impersonator

Give us feedback! React with 🚀 if perfect, 👍 if helpful, 👎 if not.