Co-Authored-By: Paperclip <noreply@paperclip.ing>
* commit '2d59dd1ba339f9ba55faaa8ae5763f4e7d8e5fb1':
fix(retention): wake blocker retention after no-run workspace export
Bound blocker workspace history reads by candidate workspace
Validate eligibility before first attention retention materialization
fix(retention): exclude finalized blockers before graph bounds
test: hold lifecycle verification before retry identity assertions
Exercise saved legacy input through its eligibility producer
fix: cancel merge confirmation acceptance under lock
Co-Authored-By: Paperclip <noreply@paperclip.ing>
* commit '74970cc21e2e5e1881516228bc7c7dbbba881a63':
fix(retention): wake blocker retention after no-run workspace export
Bound blocker workspace history reads by candidate workspace
Validate eligibility before first attention retention materialization
fix(retention): exclude finalized blockers before graph bounds
test: hold lifecycle verification before retry identity assertions
Exercise saved legacy input through its eligibility producer
Co-Authored-By: Paperclip <noreply@paperclip.ing>
* commit '5726ecf7a39339c2d68b094d9bbe1b1d0c436845':
fix(retention): wake blocker retention after no-run workspace export
Bound blocker workspace history reads by candidate workspace
Validate eligibility before first attention retention materialization
fix(retention): exclude finalized blockers before graph bounds
test: hold lifecycle verification before retry identity assertions
Exercise saved legacy input through its eligibility producer
Register finalization intent before state writes, preserving caller transactions. Shared workspace export invalidates only the company blocker source; known issue-only export stays scoped and unattached operations do not activate retention. Add actual recorder controls and the shared-root integration journey.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Share finalized and legacy operation facts across blocker pairs, retaining attributed operation precedence and strict later-finalize semantics. Qualify retained and unrelated history using actual query plans and structural work limits.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Serialize binding admission with canonical company deletion without changing the HTTP fence or current grant and revision checks. Cover both operation orders and late revocation, replacement debt, and rollback against PostgreSQL.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
* commit '6a0f5ead44362af9b8b428f49f1d43fab67808d6':
test: align deferred recovery clocks with real execution
Keep recovery acceptance timers on their owning clock
test: measure evidence retry backoff from inspection time
fix(server): recheck reactive sources after status summary commit
test: measure retention page fairness on a controlled clock
fix(heartbeat): resume saved chat work on feature enablement
test: await resolved interaction execution before outcome assertions
Wake saved conversation assignments when Agent Chat becomes enabled
* commit '1ac6ba91d6d819a801a60a876747dcd1e3c0d675':
test: align deferred recovery clocks with real execution
Keep recovery acceptance timers on their owning clock
fix(server): recheck reactive sources after status summary commit
test: measure retention page fairness on a controlled clock
fix(heartbeat): resume saved chat work on feature enablement
test: await resolved interaction execution before outcome assertions
Wake saved conversation assignments when Agent Chat becomes enabled
* commit 'cb2e694c7c28f93f08d353d7173e363e4eb2b34f':
test: align deferred recovery clocks with real execution
Keep recovery acceptance timers on their owning clock
fix(server): recheck reactive sources after status summary commit
test: measure retention page fairness on a controlled clock
fix(heartbeat): resume saved chat work on feature enablement
test: await resolved interaction execution before outcome assertions
Wake saved conversation assignments when Agent Chat becomes enabled
# Conflicts:
# server/src/services/instance-settings.ts
* commit 'b4b18c9a37188c912507e18feae9677ba1b21ffa':
test: align deferred recovery clocks with real execution
Keep recovery acceptance timers on their owning clock
fix(server): recheck reactive sources after status summary commit
test: measure retention page fairness on a controlled clock
fix(heartbeat): resume saved chat work on feature enablement
test: await resolved interaction execution before outcome assertions
Wake saved conversation assignments when Agent Chat becomes enabled
* commit '96b516dcb638cede15dae7d6488c716c5a4df18f':
test: align deferred recovery clocks with real execution
Keep recovery acceptance timers on their owning clock
fix(server): recheck reactive sources after status summary commit
test: measure retention page fairness on a controlled clock
fix(heartbeat): resume saved chat work on feature enablement
test: await resolved interaction execution before outcome assertions
Wake saved conversation assignments when Agent Chat becomes enabled
* commit '499afce285c429f814f5cd22fd437bafabf530e7':
test: align deferred recovery clocks with real execution
Keep recovery acceptance timers on their owning clock
fix(server): recheck reactive sources after status summary commit
test: measure retention page fairness on a controlled clock
fix(heartbeat): resume saved chat work on feature enablement
test: await resolved interaction execution before outcome assertions
Wake saved conversation assignments when Agent Chat becomes enabled
Keep manual receipt cursor progress when chat is disabled. Follow saved cleanup deadlines after actual child-process and PostgreSQL work settles, and preserve the exact failure-backoff boundary.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Freeze the two evidence retry fixtures and record the actual busy or failed directory inspection instant. Retain a controlled 50ms assertion lag and the full one-hour backoff requirement, expiry, reactivation, and idle checks.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Keep virtual time fixed while actual filesystem and database work settles, then advance the coordinator deadline. Preserve the three-second page-gap limit and snapshot grace assertions; starvation mutation remains detected.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Observe the target adapter-start event, then await its existing run-specific completion barrier. Preserve succeeded, execution-count and review-promotion assertions, including deliberately slow completion beyond the old terminal poll.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
