Adds the locked plan for a paid trial: does /ship cost less and catch as much when the pull request's own thread runs it, instead of a separate subagent? Why: that subagent's own steps cost about $22 of a $35 /ship run, and the keep-or-drop decision is written down here before anything is spent. Nothing to do: no code changes; paid runs start only after this merges and the budget is approved. Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> Review-Finding: p1/outside-voice#1 reader=outside-voice
CHANGELOG fragments
A PR adds its CHANGELOG entry here as changelog.d/<branch-slug>.md: the entry text only, with no ## [x] heading, and bump: minor or bump: major front matter only when the change warrants more than a patch, and important: security|regression|data-loss|blocker only when it must reach stable within about a day (docs/releases.md#the-daily-check). It never edits VERSION, package.json's version or a released CHANGELOG section. The merge step (gstack-ship-queue land, or train land in a stamped train) assigns the version (under merge: direct, the release PR from gstack-release-stamp open does), folds every fragment into one CHANGELOG section and deletes the files. The entry's first sentence (or bold lead) becomes its line in the GitHub release notes, so write it for someone using gstack: docs/releases.md#release-notes. This README is skipped. See docs/ship-policy.md#changelog-fragments.