AgentXchain v2.155.54
v2.155.54 fixes BUG-100: full-auto continuous mode now recovers from a productive local_cli turn that was killed by the hard deadline after producing output but before writing a staged result. Instead of requiring operator-side unblock, AgentXchain reissues one framework-owned retry for the same run with a 60-minute deadline.
Bug Fixes
- BUG-100: auto-retry productive timeout blockers. Continuous startup detects retries-exhausted dispatch failures with stdout evidence,
code 143or timeout diagnostics, and no.agentxchain/staging/<turn>/turn-result.json, then reissues the turn once. - Fail-closed retry budget. Silent/no-output timeout failures remain blocked, and a run gets only one productive-timeout auto-retry before emitting
productive_timeout_retry_exhausted. - Auditability. New run events
auto_retried_productive_timeoutandproductive_timeout_retry_exhaustedmake the recovery visible without changing BUG-61 ghost-blocker retry semantics.
Evidence
- npm test -- --test-timeout=60000 -> 7298 tests / 1477 suites / 0 failures / 5 skipped
Tester Re-Run Contract
Verify: npx --yes -p [email protected] -c "agentxchain --version"
For BUG-100: resume the blocked tusq.dev continuous session on the shipped package. The session must detect the retries-exhausted dev timeout blocker, emit auto_retried_productive_timeout, clear the human escalation blocker, and assign a replacement dev turn without operator-side unblock, reissue-turn, staging JSON edits, or gate mutation. If the replacement turn also exhausts the productive-timeout retry budget, the session must remain fail-closed with productive_timeout_retry_exhausted evidence instead of entering an infinite retry loop.
For BUG-99 carry-forward: pre-acceptance gate_semantic_coverage must still evaluate workflow-kit ownership with accepted turn history.
For BUG-98 carry-forward: retained QA skip-forward requests from implementation to launch must still normalize to immediate next phase qa, with stale proposed_next_role aligned only when routing-safe.
For BUG-97 carry-forward: retained active-turn run_id drift must still auto-normalize only when the staged turn_id proves active-turn ownership.
For BUG-96 carry-forward: decision entries that include existing decision text but omit rationale must still auto-normalize before schema validation.
For BUG-95 carry-forward: staged results that use files_modified instead of files_changed and omit runtime_id, summary, artifact, and proposed_next_role must still auto-normalize before schema validation.
For review-only carry-forward: a review_only role with missing or empty objections must still fail protocol challenge enforcement.
Carry-forward tester checks for shipped-package closure proof:
- BUG-54 startup watchdog evidence should quote
startup_latency_ms,elapsed_since_spawn_ms, andruntimes.<id>.startup_watchdog_ms. - BUG-55/87 verification-output evidence should quote
undeclared_verification_outputs,verification.produced_files, andgit status --shortafteraccept-turnfollowed bycheckpoint-turn. Clean tree means BUG-55 is fixed for your reproduction. - BUG-52/53 continuous recovery evidence should quote
phase_entered,reconciled_before_dispatch, andsession_continuation.