Warm recordings, file-journey-bug tasks, and run-recorded-mcp#40
Open
bhackett1024 wants to merge 5 commits into
Open
Warm recordings, file-journey-bug tasks, and run-recorded-mcp#40bhackett1024 wants to merge 5 commits into
bhackett1024 wants to merge 5 commits into
Conversation
…-mcp - tasks now carry recording_id + recording_warmed_at (+ no_retry). A recording- operating task (polish / file-journey-bug) fires warm-recording-background, which drives RecordingOverview up to 20 min: success stamps warmed_at on the recording's queued tasks; timeout/error fails them (non-retryable) and records a Replay MCP error for admin activity. - claimNextTask prioritizes dispatch by warm readiness: warmed recording first, no recording next, not-yet-warmed recording last. Used by polish tasks. - FAST MODE journeys no longer file bugs: their runner reports bug_seen + a bug_data blob into setup_output; saveTestRunSetupResult then schedules a file-journey-bug task (associated with the journey's recording, linked from the run's task page) that files the bug from the recording with no browser. - Generalized the polish setup harness into scripts/run-recorded-mcp.ts with two modes (polish, file-journey-bug); both operate on a recording via the Replay MCP without a browser. Removed scripts/polish/run-recorded.ts. - container-events defers a fast-mode journey run's finalization — and a journey eval's result callback — until the file-journey-bug task completes, so the bug is filed before the run finalizes / the eval verdict is reported. - Seeded file-journey-bug-script guidance stub; updated journey-run-script guidance context for the new FAST MODE report-a-signal behavior. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Replay QA✅ QA run passed for
|
|
🔎 Loop QA preview (all-k8s backend): https://pr-40--loop-qa-j63k5x.netlify.app |
…ecture Add docs/backend/warm-recordings.md (the dedicated architecture doc, alongside run-recorded-browse.md) covering warming, run-recorded-mcp's two modes, the file-journey-bug task, and journey-eval second-task ordering. Cross-reference it from tasks-and-containers.md, journeys.md, admin-run-journey-eval.md, test-runs.md, and run-recorded-browse.md; fix the stale journey harness path. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
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
1 participant
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.
Summary
Adds warm-recording tracking to QA tasks, splits FAST MODE journey bug-filing into a second-stage task, and generalizes the polish setup harness.
Warm recordings
tasksnow carryrecording_id,recording_warmed_at, andno_retry.warm-recording-backgrounddrivesRecordingOverviewand waits up to 20 min. On success it stampsrecording_warmed_aton the recording's queued tasks; on timeout/error it fails them (non-retryable) and records a Replay MCP error so the failure surfaces in admin activity.claimNextTaskprioritizes dispatch by warm readiness: (1) warmed-recording tasks, (2) no-recording tasks, (3) not-yet-warmed-recording tasks (last).FAST MODE journeys report a bug signal; a second task files it
bug_seen: trueplus abug_datablob (description of what it encountered + browser history) in itssetup_output.saveTestRunSetupResultthen schedules one file-journey-bug task, associated with the journey's recording and linked from the run's task page (test_run.file_journey_bug_task_id), which files the bug from the recording.run-recorded-mcp
scripts/polish/run-recorded.tsintoscripts/run-recorded-mcp.tswith two modes (polish,file-journey-bug) — both operate on a recording via the Replay MCP with no browser, mirroringrun-recorded-browse.ts.Journey eval ordering
container-eventsdefers a fast-mode journey run's finalization — and a journey eval's result callback — until the file-journey-bug task completes, so the bug is filed before the run is finalized / the eval verdict is reported.Guidance
file-journey-bug-scriptstub and updatedjourney-run-scriptguidance context for the new report-a-signal behavior. Real script content is authored in the prod DB via the guidance API.🤖 Generated with Claude Code