Conversation
…ecovery routes to inreview Three critical bugs fixed: 1. checkMergedPRsAndUpdateTasks() ignored 'blocked' tasks entirely — merged-PR detection only covered todo/inprogress/inreview. Now includes blocked tasks in mergedRecoveryCandidates and allowsMergedRecovery. 2. safeRecoverTask() always routed recovered tasks to 'todo' regardless of whether they had a linked PR. Now checks for prNumber/prUrl and routes to 'inreview' when a PR exists. 3. recoverTimedBlockedWorkflowTasks() and recoverAutoBlockedTasks() both hardcoded 'todo' as the recovery status. Now check for linked PRs and recover to 'inreview' when appropriate. Live store impact: 8 blocked tasks with merged PRs moved to done, 16 blocked tasks without PRs unblocked to todo. Zero blocked tasks remain.
- Added header scope label to the dashboard for better context on task counts. - Updated task detail fetching to include workflow run information. - Improved task classification for blocked tasks based on worktree failures. - Adjusted batch size for Vitest runner on Windows to optimize performance. - Enhanced PR fix templates to validate PR state before cloning and releasing claims. - Introduced new validation node for PR state in workflow templates. - Updated workflow nodes to support provider and provider configuration for OpenCode executor runs. - Refactored executor resolution logic to include provider and provider configuration. Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
…dex-self-improvement-loop-commits' and 'bosun/codex-self-improvement-loop-commits' of https://github.com/virtengine/bosun into bosun/codex-self-improvement-loop-commits Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
|
@copilot 2 failing tests remain: Test Files 2 failed | 260 passed | 1 skipped (263) Error: AssertionError: expected '/* Auto-generated by tools/generate-d…' to be '/* Auto-generated by tools/generate-d…' // Object.is equality
@@ -49894,19 +49894,19 @@
Error: AssertionError: expected '/* ──────────────────────────────────…' to contain 'return status === "active" || status …'
|
…nerate demo-defaults Agent-Logs-Url: https://github.com/virtengine/bosun/sessions/61d2cb29-29a9-45db-8a9e-6a26e464ae67 Co-authored-by: jaeko44 <9289791+jaeko44@users.noreply.github.com>
... Fixed both remaining failures in commit 8593011:
|
…ithub.com/virtengine/bosun into bosun/codex-self-improvement-loop-commits Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
- Removed delegation watchdog settings and related parameters from agent phases. - Eliminated auto-fix validation remediation paths and associated nodes. - Updated task status handling to set tasks as "todo" instead of "blocked" on validation failure. - Streamlined PR creation process by simplifying the PR body generation. - Adjusted metadata versioning to reflect changes in the task lifecycle template. Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
- Updated task fetching in DispatchSection to use pagination with a limit of 200 tasks. - Added recovery strategy limit to the error recovery template and introduced a new node to load recovery strategies. - Enhanced error analysis by including reusable prior strategies in the analysis prompt. - Expanded health check template to include recent run evaluations and apply self-improvement ratchets based on the latest run outcomes. - Improved task lifecycle template with enhanced validation handling, including auto-fix attempts and logging for validation failures. - Updated workflow engine to provide clearer summaries and recommendations based on workflow completion status. Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
|
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[1/1]⎯ Test Files 1 failed | 243 passed | 1 skipped (245) Error: AssertionError: expected '/* Auto-generated by tools/generate-d…' to be '/* Auto-generated by tools/generate-d…' // Object.is equality
@@ -2426,12 +2426,12 @@
@@ -2531,60 +2531,31 @@
Error: Process completed with exit code 1. |
…eCount change Agent-Logs-Url: https://github.com/virtengine/bosun/sessions/ccc0bd76-ba83-4b6d-a31d-019c686efb4b Co-authored-by: jaeko44 <9289791+jaeko44@users.noreply.github.com>
Fixed in commit 1429ad6. The workflow template was updated (the |
…sages by falling back to sendDirect test(telegram): add test for editDirect fallback when message to edit is not found test(workflow): add tests for condition.expression resolving template placeholders as JS literals test(workflow-templates): verify recovery of blocked worktrees summary uses precomputed count variable fix(ui): update demo defaults for node and edge counts, and enhance PR body configuration feat(workflow-templates): implement detailed validation failure handling with retries and notifications fix(workflow-nodes): improve condition.expression execution to prevent unsupported syntax and enhance error handling Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
…ithub.com/virtengine/bosun into bosun/codex-self-improvement-loop-commits Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
…lates with dynamic environment variables Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
|
⎯⎯⎯⎯⎯⎯⎯ Failed Tests 6 ⎯⎯⎯⎯⎯⎯⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: emits parse-safe summary with mixed telemetry quality ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[1/6]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: emits artifact and survives degraded telemetry ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[2/6]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: mixed degraded + healthy sources produce parse-safe trend deltas ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[3/6]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: partially parsed task telemetry still computes best-effort throughput ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[4/6]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: non-json task telemetry is marked degraded and suppressed from trend alerts ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[5/6]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: parses wrapped canonical task payloads and preserves planner artifact ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[6/6]⎯ Test Files 1 failed (1) JSON report written to /home/runner/work/bosun/bosun/test-results-n22-s4.json |
|
❯ tests/demo-defaults-sync.test.mjs:20:41 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[1/8]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: all templates execute without engine errors > template-weekly-fitness-summary ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[2/8]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: emits parse-safe summary with mixed telemetry quality ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[3/8]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: emits artifact and survives degraded telemetry ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[4/8]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: mixed degraded + healthy sources produce parse-safe trend deltas ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[5/8]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: partially parsed task telemetry still computes best-effort throughput ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[6/8]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: non-json task telemetry is marked degraded and suppressed from trend alerts ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[7/8]⎯ FAIL tests/workflow-guaranteed.test.mjs > guaranteed: behavioral contracts > template-weekly-fitness-summary: parses wrapped canonical task payloads and preserves planner artifact ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[8/8]⎯ Test Files 2 failed | 242 passed | 1 skipped (245) Error: AssertionError: expected '/* Auto-generated by tools/generate-d…' to be '/* Auto-generated by tools/generate-d…' // Object.is equality
@@ -2426,12 +2426,12 @@
@@ -2538,11 +2538,11 @@
@@ -2553,45 +2553,10 @@ Error: Error: template-weekly-fitness-summary: 1 engine-level error(s): Error: Error: template-weekly-fitness-summary: 1 engine-level error(s): Error: Error: template-weekly-fitness-summary: 1 engine-level error(s): Error: Error: template-weekly-fitness-summary: 1 engine-level error(s): Error: Error: template-weekly-fitness-summary: 1 engine-level error(s): Error: Error: template-weekly-fitness-summary: 1 engine-level error(s): Error: Error: template-weekly-fitness-summary: 1 engine-level error(s): |
…(\\w → \w) Agent-Logs-Url: https://github.com/virtengine/bosun/sessions/00be45d1-d79a-4c02-bee1-f6cfce28c9a8 Co-authored-by: jaeko44 <9289791+jaeko44@users.noreply.github.com>
Fixed in commit 01f22cf. The |
- Updated the workflow completion summary to provide more context when there are pending nodes on inactive branches. - Improved messaging for terminal completed states to clarify when no further action is required. - Introduced a new function `collectTaskBranchBindingCandidates` to gather potential branch candidates from task metadata. - Enhanced the `trigger.task_available` node to log task branch binding details, including resolved branches and candidates. - Refined regex patterns in condition expressions to improve security by preventing unsupported syntax. Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
…ithub.com/virtengine/bosun into bosun/codex-self-improvement-loop-commits Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
… runtime metrics Co-authored-by: bosun-ve[bot] <262908237+bosun-ve[bot]@users.noreply.github.com>
No description provided.