Pull Request Explorer

Summary

Fix Playwright extension dry-run parsing

Ensures Playwright build extension correctly parses new dry‑run output, preventing build failures and ensuring accurate browser installation paths.

Health Assessment

• • Quick fix with minimal iterations, resolved within 12.6 hours, indicating efficient review and low complexity.

Summary

Remove obsolete task event indexes

Eliminates redundant ClickHouse indexes to reduce storage overhead and improve performance.

Health Assessment

• • PR was reviewed and merged within 2.3 hours, indicating minimal friction and low risk.

Summary

chore: release v4.5.16

Releases version 4.5.16, updating changelogs and package versions across packages, ensuring consistent dependency updates and release notes.

Health Assessment

• • The PR was merged quickly with a single automated commit and a single comment review, indicating a smooth, low‑friction release process.

Summary

Add infrastructure error classifier and retry helper

Provides utilities to detect transient database connectivity errors and automatically retry safe operations, improving system resilience and reducing user-facing failures.

Health Assessment

• • The PR was reviewed quickly by AI reviewers, leading to minimal back‑and‑forth and a single commit. The large code addition was handled efficiently, indicating a well‑structured change.

Summary

Add DB blip harness for connection-resilience tests

Provides a test harness that simulates Postgres connection blips, enabling tests to verify database code resilience to dropped connections.

Health Assessment

• • Fast review and single commit indicate a smooth process; the large test harness addition enhances reliability testing for database interactions.

Summary

Fix CLI code layer build to use exclude copy

Improves deployment speed by avoiding deletion of node_modules during Docker image build, reducing build time and resource usage.

Health Assessment

• • Quick fix with minimal changes, AI review expedited the process.

Summary

Health Assessment

Summary

Fix run-ops sharding failures and 404 unroutable IDs

Improves observability of run-ops sharding errors and ensures unroutable IDs return 404, reducing silent failures and improving monitoring.

Health Assessment

• • Quick turnaround with minimal review iterations; large code changes but well-reviewed.

Summary

Prepare task event attribute inserts for ClickHouse

Optimizes ClickHouse writes by explicitly naming insert columns, ensuring future schema changes won’t break attribute handling. This change maintains backward compatibility while preparing for a later storage schema update.

Health Assessment

• • Quick review with minimal changes indicates a straightforward performance tweak.

Summary

chore: ignore IDE settings

Prevents local IDE settings from appearing as untracked repository changes.

Health Assessment

• • Quick, single-commit PR with no review friction, minimal scope.

Summary

Improve Vercel & Slack integration settings pages

Enhances the user experience by aligning the Vercel and Slack integration settings UI with the rest of the settings pages, adding missing navigation, and improving accessibility.

Health Assessment

• • The PR received a rapid review but required several subsequent commits, indicating iterative UI refinement. The large line change across only four files suggests a focused but substantial update, resulting in moderate risk.

Summary

Ignore Vim swap files

Prevents local Vim swap files from appearing as untracked repository changes.

Health Assessment

• • Quick change with no rework, merged within 0.7 hours.

Summary

Verify public-to-private repo sync

Adds a temporary marker used to verify repository synchronization.

Health Assessment

• • Rapid review with no issues; minimal changes and quick merge.

Summary

chore: establish repository synchronization baseline

Adds a baseline marker to establish the initial synchronization boundary for repository operations, ensuring consistent state across environments.

Health Assessment

• • PR merged quickly with no review issues, indicating minimal risk and straightforward implementation.

Summary

Add repository synchronization gates

Adds disabled repository synchronization workflows and guards merge‑queue entries against unsynchronized changes, remaining inactive until explicitly enabled.

Health Assessment

• • Fast review and minimal rework indicate a straightforward change with low risk.

Summary

Flag legacy Node deployments for runtime updates

Ensures legacy Node.js deployments are included in runtime update reports, keeping dashboard and CLI output consistent.

Health Assessment

• • Quick review and minimal changes indicate low complexity and high confidence in the fix.

Summary

Deploy build flags and config auto-detection docs

Updated deployment documentation to clarify build flags and auto-detection, improving developer onboarding.

Health Assessment

• • Quick, single-commit docs update with minimal review comments, indicating low complexity and high confidence in correctness.

Summary

Fix dead‑letter redrive channel mismatch

Ensures dead‑letter redrive messages are published to the correct queue channel, preventing silent failures and improving reliability for non‑default queues.

Health Assessment

• • Quick, single‑commit fix with AI‑assisted review; minimal code churn and no major risk.

Summary

Warn on missing createRequire packages in deployments

Adds runtime warnings for packages loaded via createRequire that will be missing from deployed images, preventing silent failures and guiding developers to add the necessary packages.

Health Assessment

• • The PR was reviewed and merged within 5 hours after a single AI‑review round, but the large number of added lines and extensive comment activity suggest a significant change with moderate risk.

Summary

Guard run finalization against lost resume signals

Ensures parent runs do not hang when resume signals are lost, improving system reliability and preventing indefinite waits for dependent jobs.

Health Assessment

• • Fast review and single commit indicate low friction; large code changes suggest moderate risk but thorough testing mitigates it.

Summary

chore: release v4.5.15

Deploys version 4.5.15 by updating changelogs, package versions, and Helm chart, while removing compiled test files and adding side channel improvements for better runtime efficiency.

Health Assessment

• • Fast review and single commit led to a smooth merge within 47 hours, indicating a well‑structured release process with minimal friction.

Summary

Update integration settings copy and toggle

Improves build settings UI, adds admin‑only toggle, and secures native build server option handling.

Health Assessment

• • Fast review and merge with minimal rework; AI review tools identified security and UI issues early.

Summary

Move Queues search and pagination above table

Repositions search and pagination controls to align with the queue table, improving UI clarity and reducing user confusion.

Health Assessment

• • Quick fix with minimal changes and a single AI review, indicating low complexity and high confidence.

Summary

Allow 10 session trigger tags

Aligns session trigger tag limit with run tag limit, improving consistency and user experience.

Health Assessment

• • Fast review cycle with AI review bots, minimal rework, and a small code change.

Summary

Stop publishing compiled test files

Removes test files from published npm packages, reducing install size and preventing unresolved dependencies.

Health Assessment

• • Fast turnaround with a single review, minimal changes, and no reported issues.

Summary

Add named side channels to Session streams

Enables durable, cross‑run realtime streams for sessions, allowing agents to stream data such as screenshots or telemetry without waking runs, improving user experience and data flow.

Health Assessment

• • Rapid review and single iteration indicate focused changes; large scope and many comments suggest significant impact and moderate complexity; AI‑assisted review helped streamline feedback.

Summary

Polish organization Projects settings page

Enhances user experience by refining UI elements, adding update counts, and improving project visibility on the organization settings page.

Health Assessment

• • Quick review and merge with minimal changes, indicating a straightforward UI polish.

Summary

chore: release v4.5.14

Release new version 4.5.14 with updated dependencies and changelog updates across packages.

Health Assessment

• • Release PR merged quickly with no review changes, indicating smooth process.

Summary

Fix task retry queue budget exhaustion

Ensures long‑delayed task retries no longer consume the queue nack budget, preventing premature dead‑lettering and improving run reliability.

Health Assessment

• • Quick, single‑commit fix with minimal review friction; AI review bots provided only minor nitpicks.

Summary

Emit workload auth gate metrics via OpenTelemetry

Adds OpenTelemetry metrics for workload authorization, improving observability and enabling accurate traffic accounting across workers.

Health Assessment

• • PR was reviewed and merged within minutes, with AI tools providing quick feedback and no rework required.

Summary

Skip publish for non-image packages

This PR prevents unnecessary Docker image builds and downstream dispatches when changes are made to packages that are not included in the webapp or worker images, reducing CI/CD noise and resource usage.

Health Assessment

• • Quick review and merge with minimal changes indicates low complexity and high confidence in the change.

Summary

Improve Usage Page Billing Panels

UI-only update of the Usage page, enhancing billing panel layout and adding billing limit link, improving user experience.

Health Assessment

• • Rapid review and merge, minimal changes, quick turnaround.

Summary

Compact build logs for native deploys

Reduces log noise during native build deployments by showing a single spinner line and the last 20 lines on failure, improving developer experience and CI clarity.

Health Assessment

• • Fast review cycle (0.1h) and short merge time (1.2h) indicate minimal friction; AI review tools helped surface issues early, leading to a single commit fix and quick merge.

Summary

Fix batch list to scan all run‑ops stores

Ensures the batches page displays all batches across every configured run‑ops store, eliminating silent data loss and improving reliability.

Health Assessment

• • Fast review and merge after a single AI‑reviewed comment, but the change spans over 500 lines across multiple stores, indicating a substantial code modification.

Summary

Polish AI agent setup panel

Improves onboarding experience for new projects by refining the AI agent setup panel, reducing friction and ensuring consistent UI.

Health Assessment

• • Quick review and merge with minimal iterations, indicating straightforward UI polish.

Summary

Health Assessment

Summary

Add start-from-latest streams and useSessionStream hook

Enables real‑time streaming from the latest record, bounded memory, resumable sessions, and token refresh for uninterrupted user experience.

Health Assessment

• • Rapid iteration with AI‑assisted code and review, resulting in a quick merge within 3 hours.

Summary

Add three-shard coverage for run-ops router tests

Enhances test coverage for run-ops router across shard configurations, ensuring routing logic behaves correctly in multi-shard environments. This prevents hidden bugs in shard routing that could affect production.

Health Assessment

• • Fast review cycle with AI-assisted review, minimal rework, and extensive test coverage expansion.

Summary

Release v4.5.13

Deploys version 4.5.13 with new features, improvements, and bug fixes across CLI, server, and SDK, enhancing stability and functionality.

Health Assessment

• • Release PR with minimal review, quick approval, but large documentation changes.

Summary

Add server-selected deploy build path

Enables the CLI to query the server for the appropriate build path, allowing per‑organization and per‑environment build path selection without requiring a new CLI release.

Health Assessment

• • Rapid iteration with multiple AI reviews; large code changes but quick cycle time indicates an efficient process.

Summary

Add runtime clientData validation for custom agents

Enables custom chat agents to validate and parse client data against a schema, improving error handling and reliability for end‑users.

Health Assessment

• • The PR required extensive rework over 10 days, with multiple review comments and many commits, indicating high complexity and risk.