Pull Request Explorer

Summary

Health Assessment

Summary

Fix AskUserQuestion UI and desktop notification

Improves user experience by correcting highlight color and ensuring macOS dock alerts on approval.

Health Assessment

• • Quick turnaround with minimal review comments, indicating straightforward UX fixes.

Summary

Add deep-review skill, superseding review-checklist

Introduces a comprehensive deep-review skill for code analysis, replacing the older review-checklist and enhancing review depth and quality.

Health Assessment

• • The PR was merged within 13 hours with no human review, relying on automated AI review comments; the change is purely documentation, so risk is minimal and the process was efficient.

Summary

Limit memory extraction concurrency

Reduces database load and improves performance by throttling concurrent memory extraction workflows.

Health Assessment

• • Fast cycle time and minimal review iterations indicate smooth integration.

Summary

Fix worktree removal and skill drag preview

Improves chat input UX by enabling branch worktree deletion, correcting skill drag icon, and preventing accidental AI autocomplete submissions.

Health Assessment

• • Fast cycle time of 0.5h, single review with no rework, medium scope of 13 files and 64 lines, indicating low risk and smooth integration.

Summary

Fix orphan assistant placeholder on tool-first resume

Removes redundant empty assistant rows created during tool-first resume flows, improving data cleanliness and user experience in share/replay views.

Health Assessment

• • Fast review cycle with AI review comments, minimal rework, moderate scope bug fix.

Summary

Reuse single verify report across re-ingests

Enables iterative verification without duplicate reports, streamlining report management and reducing clutter for users.

Health Assessment

• • Fast cycle time and minimal review rounds indicate a straightforward, low-risk change.

Summary

Polish chat UI: tighten overlay border, generic icon

Improves chat input UX by tightening drag overlay border and standardizing local device icon, enhancing visual consistency and usability.

Health Assessment

• • The PR was merged quickly with minimal review, indicating a straightforward UI polish with low risk.

Summary

Fix agent runtime Redis persistence overflow

Prevents large message payloads from exceeding Redis limits, ensuring reliable operation and data integrity.

Health Assessment

• • The PR resolved a critical Redis size issue with minimal code changes and a quick review, but the extended cycle time indicates possible scheduling or prioritization delays.

Summary

Polish UI details across chat, community, onboarding

Enhances user experience by improving visual focus cues, reducing motion distractions, and preventing accidental navigation, leading to a smoother, more accessible interface.

Health Assessment

• • The PR was reviewed immediately and merged within 21 hours, indicating a smooth process.
• • The moderate code changes and single commit suggest low risk and minimal friction.

Summary

Cap readDocument content to avoid context overflow

Prevents LLM context overflow by truncating large document content, ensuring stable agent performance.

Health Assessment

• • Fast turnaround with minimal rework; single review and quick merge indicates low complexity.

Summary

Fold non‑latest finished turns into header

Improves chat UI by collapsing older turns into a concise header, reducing clutter and enhancing readability for users.

Health Assessment

• • The PR was reviewed and merged within 36 minutes, with no rework required, indicating a smooth and low‑risk change.

Summary

Fix delete 404 and UI issues in agent documents

Improves user experience by preventing 404 errors when deleting documents, adding missing icons, and ensuring server‑side changes refresh client lists.

Health Assessment

• • Fast turnaround with AI assistance; single commit, quick review, minimal friction.

Summary

Fix floating chat panel expand offset

Adjusts the floating chat panel’s expand affordance to account for the chat input overlay height, preventing visual overlap with status and queue trays and improving UI consistency.

Health Assessment

• • Fast turnaround with minimal rework; single review comment and quick merge indicates low complexity and risk.

Summary

Health Assessment

Summary

Add floating panel for task topic chat

Enhances task topic chat UI by replacing drawer with a resizable floating panel, improving usability and reducing clutter.

Health Assessment

• • PR merged within 36 minutes with minimal review comments, indicating a straightforward change with adequate tests.

Summary

Health Assessment

Summary

Fix workspace problems

Resolve workspace-related issues to improve stability and user experience.

Health Assessment

• • Rapid iteration with minimal review; large code churn indicates significant refactoring of workspace logic.

Summary

Fix byStatus group mode rendering in sidebar

Corrects sidebar rendering for byStatus grouping, ensuring topics display their working directory and icons align properly, improving user experience.

Health Assessment

• • Fast turnaround with minimal review friction; quick fix and style adjustments resolved the issue promptly.

Summary

Fix custom MCP connectors in @ mention picker

Enables user‑added MCP connectors to be @‑mentioned and callable, improving chat tool integration and UI clarity.

Health Assessment

• • PR merged within 1.2 hours with no review rework; AI review accelerated validation and documentation.

Summary

Fix crash rendering local file tags

Prevents crashes when rendering messages containing local file tags, improving stability for users dragging files into chat.

Health Assessment

• • Quick, single-commit fix with minimal code changes and no impact on other components.

Summary

Persist CC resume token on stream_start to survive watchdog abandon

Fixes a bug where the resume token was lost when a session was abandoned, ensuring continuity of user context and preventing loss of progress.

Health Assessment

• • Fast cycle time of 2.7 hours, minimal review, quick fix to a critical bug.

Summary

Add workflow run guard for performance

Introduces a Redis-backed guard to prevent duplicate or overlapping workflow runs, enhancing reliability and resource efficiency.

Health Assessment

• • Merged within 30 minutes of opening, indicating a smooth review process and high confidence in the changes.

Summary

Fix client notification to current run

Ensures desktop notifications display the correct assistant reply in multi‑turn chats, preventing stale messages.

Health Assessment

• • Quick fix with minimal rework; automated review identified the issue and the PR was merged within 9 hours.

Summary

Add restore to input action for user messages

Adds a quick-resend feature that refills the chat input with the original message text and attachments, improving user experience.

Health Assessment

• • Fast review cycle with minimal rework, AI-assisted code generation streamlined the implementation.

Summary

Normalize skill metadata wording

Standardize skill description phrasing across the repository for consistency and clarity.

Health Assessment

• • The PR was reviewed quickly with automated tools, and the changes were minor and focused on documentation.

Summary

Apply Codecov threshold to each flag status

Adjusts Codecov CI checks to allow a 0.5% coverage tolerance per project flag, preventing false failures on minor coverage drops.

Health Assessment

• • Fast, single-commit PR with no rework, minimal risk.

Summary

Add Claude Code quota menu to chat input

Adds a read‑only subscription‑quota menu for the Claude Code heterogeneous agent, mirroring the existing Codex quota menu and improving user visibility into usage limits.

Health Assessment

• • Fast review cycle with AI‑powered review comments enabled rapid validation; minimal rework and clear scope suggest a low‑friction, medium‑risk feature addition.

Summary

Fix missing document not-found display

Ensures the entire document module displays a 404 when a requested document is missing, preventing misleading headers and improving user experience.

Health Assessment

• • Fast cycle time and minimal review comments indicate a straightforward bug fix with quick resolution.

Summary

Remove LOBE-XXX markers from code

Clean up internal tracking markers, improving code readability and security.

Health Assessment

• • Fast turnaround with minimal review, indicating straightforward cleanup.

Summary

Expose select/forward action on finished hetero agent turns

Fixes UI bug where select/forward actions were hidden after heterogenous agent turns, improving user experience.

Health Assessment

• • Quick fix with minimal rework, AI-assisted code contributed, fast review cycle.

Summary

Fold single-sentence status lines into workflow

Improves chat UI by ensuring only concise status updates are collapsed, making narration clearer and reducing hidden content for users.

Health Assessment

• • Merged within 1.1 hours with minimal review iterations, indicating high confidence and low risk.

Summary

Add remote Claude Code Human-in-the-loop

Enables structured AskUserQuestion capability in remote Claude Code, improving user interaction

Health Assessment

• • The PR has a large scope with 12 files changed, but the review process was relatively smooth with a short time to first review

Summary

Improve verify report UI interactions

Enhances the verify report UI with inline evidence previews, rename/delete actions, and hides report list for anonymous links, improving user experience and reducing friction.

Health Assessment

• • Merged within 0.8 hours with minimal review, indicating a smooth and efficient process.

Summary

Add write-side save-state framework

Improves autosave reliability by adding failure handling, preventing silent data loss and boosting user trust.

Health Assessment

• • Fast cycle time (1.6h) and immediate review indicate a smooth process; AI assistance helped reduce manual effort and the PR required minimal back‑and‑forth discussion.

Summary

Update document tree icon to manuscript glyph

Aligns document tree icons with the Pages tab, improving visual consistency and user clarity.

Health Assessment

• • The PR was reviewed immediately and merged within 45 hours with no rework, indicating a smooth, low-risk style change.

Summary

Fix dead local links in agent testing docs

Updates agent testing documentation to ensure evidence links are clickable, improving user experience.

Health Assessment

• • Quick merge with no review friction, minimal scope.

Summary

Polish Codex heterogeneous agent profile UI

Lightens Codex quota trigger text for better readability. Hides the Agent Builder panel on heterogeneous agent profiles to avoid confusion.

Health Assessment

• • Quick turnaround with minimal changes and a single review comment, indicating a smooth process.

Summary

Expose select/forward action on finished hetero agent turns

Adds missing select and forward options for heterogeneous agent messages, improving user interaction. Ensures consistent action menu across all agent types.

Health Assessment

• • Quick resolution with minimal rework; AI-assisted code generation streamlined fix.

Summary

Remove legacy TTS providers and add prototype skill

Streamlines TTS functionality by removing outdated providers and introduces a new design prototype skill, improving user experience and simplifying maintenance.

Health Assessment

• • Rapid cycle with minimal rework; AI review helped catch issues early.

Summary

Stabilize agent-scroll pin assertion with polling

Improves reliability of end‑to‑end tests for agent scroll behavior by polling for stable pin position, reducing flaky test failures.

Health Assessment

• • Quick review and merge, minimal changes, no risk to production.

Summary

Add Codex Fast speed mode to model selector

Enables a 1.5x speed mode for Codex agents, adding UI controls and CLI options to improve user experience and reduce processing time.

Health Assessment

• • Fast cycle time and single review comment indicate a straightforward, low-risk feature addition.

Summary

Add verification report library

Adds a verification report library and master‑detail workspace for users to view and manage verification reports, improving usability and reducing navigation friction.

Health Assessment

• • The PR was merged in under 14 hours with minimal review rounds, indicating a smooth process; AI review tools helped catch lint and formatting issues early.

Summary

Surface fetch failure states across read surfaces

Migrate read surfaces to show error and retry instead of permanent skeleton on fetch failure

Health Assessment

• • Multiple review rounds and commits indicate significant changes and potential for errors

Summary

Add dedicated inspector labels for node/git/python commands

Enhances the command inspector UI by distinguishing node, git, and python commands with specific labels and brand icons, reducing visual clutter and improving agent usability.

Health Assessment

• • The PR was reviewed and merged within an hour, with no rework or comments, indicating a smooth, low‑friction process.

Summary

Unify MCP issue label taxonomy and auto-label rescan requests

Centralizes MCP label names and auto-labels rescan requests, improving issue tracking and automation

Health Assessment

• • The PR was reviewed and merged quickly, indicating a smooth development process
• • The changes are well-structured and include tests, reducing the risk of errors

Summary

Drag workspace files into chat input

Enables users to drag files or folders from the workspace sidebar into the chat input, inserting a lightweight local file reference instead of uploading a copy, improving workflow efficiency.

Health Assessment

• • Fast review and merge with minimal iteration, but large code change may increase maintenance risk.

Summary

Fix breadcrumb route prefixes for agent chat

Ensures agent breadcrumb links maintain workspace context, preventing navigation errors.

Health Assessment

• • Quick fix with minimal changes, AI-assisted review helped catch issues early.

Summary

Fix retry logic for content filter errors

Prevents unnecessary retries when content policy violations occur, improving reliability and reducing costs.

Health Assessment

• • Quick fix with minimal changes, reviewed immediately, no rework needed.

Summary

Fix memory workflow cursor serialization

Prevents crashes when restoring workflow cursors, ensuring reliable user data processing.

Health Assessment

• • Fast cycle time and minimal review indicate a straightforward bug fix with quick resolution.