Repository navigation
chore(upstream): track the snooze fix for threads snoozed during their own run - #771
Merged
Merged
Conversation
Upstream divergenceBoth sides are compared with upstream at the in-order tip
No upstream file changes how much it differs from upstream. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Upstream pingdotgg#17118 fixes a snooze that's lost when it's set during the thread's own run. For example, an agent that snoozes its own thread through
t3_thread_organizebefore its final reply wakes the thread as soon as that reply completes. The fork has the same bug.The PR doesn't apply to fork
main: it builds on pingdotgg#16782 (isSnoozed), which the queue'searlyplanner can't import because it conflicts with a chain of pending MCP changes. This adds a watch entry to.github/upstream-tracked-prs.jsonso the PR shows up in the tracked report and is taken at chronological intake. The entry also notes the related open issue pingdotgg#15135, which the PR doesn't cover.Validation:
node scripts/upstream/tracked-prs-report.ts --fork-ref HEADlists pingdotgg#17118 as open with this reason.Written by an agent (Claude Code, claude-opus-5-5).