Pull Request Explorer

Summary

Block package export when sub-workflow missing

Prevents exporting packages that reference missing sub-workflows, ensuring export integrity and reducing runtime errors for users.

Health Assessment

• • The PR had a long cycle time and slow first review, but all changes were made before the review, indicating a large, complex change that required extensive testing and validation.

Summary

Fix AI test run stats classification

Ensures AI-generated workflow tests are counted as manual executions, preventing skewed production metrics and user milestones.

Health Assessment

• • Quick backport with minimal changes and fast review, indicating low complexity and high confidence in correctness.

Summary

Add dummy get-n8n.sh script

This script enables testing of the new get.n8n routing, ensuring routing logic works before deployment.

Health Assessment

• • PR merged in minutes with no review or comments, indicating minimal complexity and low risk.

Summary

Health Assessment

Summary

Fix external module resolution in Code node

Ensures Code node can load allowed external modules in Docker, preventing runtime errors and improving reliability for users.

Health Assessment

• • Quick fix with minimal changes, resolved within 20 hours, low risk to production.

Summary

Fix AI test run stats classification

Ensures AI-generated workflow tests are counted as manual executions, preventing skewed production metrics and user milestones.

Health Assessment

• • Rapid backport with minimal review, indicating a straightforward fix that addressed a metric misclassification.

Summary

Export CommandBarItem type from design system

Adds a top‑level re‑export of CommandBarItem, allowing consumer packages to import the type without deep paths and enabling the module SDK to reference the canonical type.

Health Assessment

• • Quick review and single commit led to a fast merge; AI assistance streamlined the change with minimal friction.

Summary

Fix AI Assistant empty view scrolling

Enables vertical scrolling in the AI Assistant empty view when content overflows, improving usability on smaller screens.

Health Assessment

• • Quick backport with minimal changes, no review delays, indicating low complexity and high confidence.

Summary

Fix AI Assistant empty view scrolling

Enables vertical scrolling in the AI Assistant empty view when content overflows, improving usability on smaller screens.

Health Assessment

• • Quick, single-file change with no tests added; fast review and merge cycle indicates low complexity and minimal risk.

Summary

Fix uneven spacing between workflow metadata dividers

Corrects visual layout of workflow card metadata by making divider characters standalone flex children, ensuring consistent spacing.

Health Assessment

• • Rapid resolution with minimal changes; test added to guard against regressions.

Summary

Count AI test runs as manual executions

Ensures AI-generated workflow tests are recorded as manual runs, preventing skewed production metrics and user milestones.

Health Assessment

• • Fast review and merge with minimal rework; medium code changes; low risk to production.

Summary

Add sign-in with person and app credentials

Enables Microsoft Excel SharePoint node to authenticate using either user or app credentials, expanding automation capabilities for unattended workflows.

Health Assessment

• • Fast review and merge with minimal rework, indicating clear implementation and thorough testing.

Summary

Fix AI Assistant empty view scrolling

Ensures users can access all suggestion options on smaller screens, improving usability and preventing content from being clipped.

Health Assessment

• • Quick, single-file change with minimal review and no tests added; fast cycle time indicates low complexity and risk.

Summary

Add Instance AI durable event-log store

Provides durable persistence for AI event logs, enabling reliable replay, analytics, and improved system reliability without affecting current runtime behavior.

Health Assessment

• • Long cycle time (120h) and multiple review rounds indicate significant review friction; large scope and many files suggest high risk for integration and testing.

Summary

Fix default MCP tool transport to httpStreamable

Ensures new agents use the correct, non-deprecated transport, improving reliability and preventing potential communication failures.

Health Assessment

• • Quick resolution with minimal rework; the change fixed a specific bug with a small code footprint and was merged within 20 hours.

Summary

Sort public API tags alphabetically

Reorders public API tags for better documentation rendering

Health Assessment

• • Quick review and merge, low complexity changes

Summary

Switch CI eval test source to LangTracer

This PR updates CI eval runners to fetch test cases from LangTracer, removing local JSON files and ensuring the test suite stays current.

Health Assessment

• • Fast review and minimal rework; small scope but long cycle time due to merge timing.

Summary

Fix AI assistant quota exhaustion error states

Clarifies out‑of‑credits errors for AI assistant, improving user experience with clear messages and upgrade prompts.

Health Assessment

• • Rapid backport with minimal review, indicating high confidence in the changes.

Summary

Bind dynamic credential tokens to user ID

Ensures dynamic credential URLs are tied to the initiating user, preventing unauthorized credential activation and enhancing security.

Health Assessment

• • Rapid backport with minimal review, indicating high confidence in the change.

Summary

Migrate backend packages to TypeScript 7

Upgrades core backend modules to TypeScript 7, enhancing type safety and maintainability for future development.

Health Assessment

• • The PR was merged within 30 minutes of opening, with a single AI-assisted review and no additional comments, indicating a smooth and efficient process.

Summary

Make AI assistant quota-exhausted error states explicit

Provides a clear out‑of‑credits notification for AI Assistant users, improving user experience and encouraging plan upgrades.

Health Assessment

• • The PR spanned nearly 140 hours with multiple commits after the initial review, reflecting a complex refactor across backend, frontend, and test layers. The large scope and extended cycle time suggest a high risk of integration issues, mitigated by extensive automated tests and AI-assisted code generation.

Summary

Bind dynamic credential tokens to a user ID

Ensures dynamic credential URLs are tied to the initiating user, preventing unauthorized activation and improving security.

Health Assessment

• • Rapid review and merge within 20 hours, minimal rework, but large code changes addressing a critical security issue.

Summary

Bound and instrument Instance AI eval lane containers

Hardens CI to prevent memory and disk failures during AI evaluation runs, ensuring reliable and diagnosable job execution for production stability.

Health Assessment

• • The PR required many iterative commits to address CI hardening, resulting in a slow cycle time but minimal review friction.

Summary

Wire Schedule Trigger node into durable scheduler

Enables persistent scheduling of trigger rules in the database, improving reliability and scalability across instances while keeping the node engine‑agnostic.

Health Assessment

• • The PR received extensive automated review comments but was resolved in a single review round, indicating efficient collaboration and a moderate level of review friction.

Summary

Extract module contract and registries into frontend SDK

Improves modularity and decouples module contracts from editor UI, enabling easier maintenance and potential reuse without affecting runtime behavior.

Health Assessment

• • Single review cycle with quick approval and no major rework, indicating a clean refactor with minimal risk.

Summary

Add API endpoints for evaluation test runs

Enables programmatic triggering and cancellation of evaluation test runs, improving automation and control for users.

Health Assessment

• • The PR was reviewed quickly with minimal comments and required only two additional commits, indicating a smooth review process. The change is moderately large (≈427 lines across 17 files) but does not involve high-risk production fixes.

Summary

Import workflows from nested package folders

Adds the ability to import workflows that are nested inside package folders into their correct location, improving import workflow handling and reducing manual steps.

Health Assessment

• • The PR required over three days to merge, with a slow review process and multiple commits after the first review, indicating moderate complexity and potential friction.

Summary

Polish custom instance role UX and copy

Improves clarity and correctness of role permissions, reducing user confusion and preventing accidental privilege changes.

Health Assessment

• • Quick review and merge with minimal iterations, indicating high confidence in changes.

Summary

Add one-tap in-chat approval for Telegram

Enables approvers to approve or decline directly within Telegram chat, improving workflow speed and reducing friction.

Health Assessment

• • The PR required multiple iterations after the initial review, indicating complex changes and a lengthy review cycle.

Summary

Add hidden Microsoft Excel SharePoint node shell

Introduces a placeholder node for future Excel/SharePoint integration, keeping existing workflows functional while the feature is under development.

Health Assessment

• • The PR was reviewed and merged within 18 hours, with a single commit and minimal comments, indicating a smooth, low-risk change.

Summary

Fix flaky next/prev button behavior in editor

Resolved flakiness in E2E tests by preventing the inline expression editor's navigation buttons from being disabled due to cursor hover, ensuring consistent navigation in the Node Details View editor and improving test reliability.

Health Assessment

• • M
• • i
• • n
• • i
• • m
• • a
• • l
• •
• • c
• • o
• • d
• • e
• •
• • c
• • h
• • a
• • n
• • g
• • e
• • s
• •
• • a
• • n
• • d
• •
• • a
• •
• • s
• • i
• • n
• • g
• • l
• • e
• •
• • q
• • u
• • i
• • c
• • k
• •
• • r
• • e
• • v
• • i
• • e
• • w
• •
• • i
• • n
• • d
• • i
• • c
• • a
• • t
• • e
• •
• • l
• • o
• • w
• •
• • c
• • o
• • m
• • p
• • l
• • e
• • x
• • i
• • t
• • y
• •
• • a
• • n
• • d
• •
• • r
• • i
• • s
• • k
• • ,
• •
• • w
• • i
• • t
• • h
• •
• • t
• • h
• • e
• •
• • f
• • i
• • x
• •
• • d
• • i
• • r
• • e
• • c
• • t
• • l
• • y
• •
• • a
• • d
• • d
• • r
• • e
• • s
• • s
• • i
• • n
• • g
• •
• • t
• • e
• • s
• • t
• •
• • s
• • t
• • a
• • b
• • i
• • l
• • i
• • t
• • y
• • .

Summary

Add repo-local Nathan CLI for test deployments

Enables developers to spin up test instances locally via a CLI, reducing reliance on Slack and speeding up deployment workflows.

Health Assessment

• • The PR added a new CLI tool for deploying test instances, required several iterations and review rounds, and was merged after roughly four days of development.

Summary

Add Timeout option to AWS Bedrock Chat Node

Enables configurable request timeouts and retry logic for AWS Bedrock Chat Model, preventing indefinite hangs and improving reliability for AI workflows.

Health Assessment

• • Fast turnaround with minimal rework; single review and few commits indicate a straightforward change.

Summary

Ping release captains when release fails

Notify release captains via CI when a release fails, improving incident response.

Health Assessment

• • Quick merge with minimal changes and no review friction.

Summary

Prevent sub-workflow failures after user transfer

Fixes intermittent sub-workflow failures caused by stale ownership cache when transferring deleted users, ensuring reliable workflow execution and preventing user-facing errors.

Health Assessment

• • The PR addressed a critical bug affecting workflow execution after user deletion, but the review and merge process took over 6 days, indicating potential bottlenecks in the review workflow.

Summary

Gate Instance AI config evals behind feature flag

Adds a PostHog feature flag to control Instance AI config‑based evaluation tools, enabling per‑user gating and preserving the skill catalog cache. This limits exposure of new AI capabilities until the feature is fully ready.

Health Assessment

• • Rapid iteration with AI assistance, quick review, and minimal rework; feature added with per‑user gating and cache preservation.

Summary

Release 2.31.0

This release updates dependencies and documentation, delivering new features and bug fixes across the platform.

Health Assessment

• • Rapid merge with minimal review indicates a stable release process.

Summary

Stabilize executions filter reset e2e test

Fixes flaky Playwright test for executions filter reset, improving test reliability and reducing false failures in the automation suite. This ensures consistent behavior of the executions list UI, giving developers confidence in the feature.

Health Assessment

Summary

Add N8N_MCP_BASE_URL for split-hostname MCP OAuth

Enables self-hosted instances with separate MCP hostnames to authenticate AI connectors without OAuth failures, improving reliability for remote AI integrations.

Health Assessment

• • PR resolved a critical OAuth issue for split-hostname deployments, required a single review and a few commits, but the 82‑hour cycle time indicates moderate complexity and a slower merge process.

Summary

Limit end-user credential creation to privileged roles

Restricts creation of end-user credentials to owners, admins, and project admins, enhancing security by preventing unauthorized credential management.

Health Assessment

• • Fast review and merge with minimal rework; single review round; comprehensive tests added.

Summary

Rework builder interactive tools onto shared contract

Enables consistent interaction contracts across agents and instance AI, improving reliability and reducing duplicated code, enhancing user experience.

Health Assessment

• • Fast review and merge cycle, but large scope of changes suggests high complexity; single review round indicates confidence in code quality.

Summary

Add Instance AI thread launch from templates

Enables users to start AI assistant threads directly from workflow templates on the website and instance template page, improving onboarding and customization.

Health Assessment

• • The PR had a long cycle time (≈172 h) and required many commits after the first review, indicating significant rework. The large scope (≈815 lines across 20 files) and multiple review rounds suggest high complexity and risk.

Summary

Improve node group selection UX

Enhances workflow editor by making node groups fully selectable, adding context‑menu and keyboard support, improving efficiency and reducing user errors.

Health Assessment

• • PR merged after 4+ days with multiple iterations post‑review, indicating moderate review friction and a large scope of changes.

Summary

Remove preview label from MCP settings

Eliminates the preview tag from the MCP settings UI, simplifying the interface now that the feature is generally available.

Health Assessment

• • Fast review cycle, minimal changes, AI-assisted code generation, low complexity and risk.

Summary

Add build-agent orchestration tool for instance AI

Enables AI assistants to delegate agent building to a non-interactive sub-agent, improving workflow automation and user experience.

Health Assessment

• • Merged in under 6 hours with minimal review friction, indicating strong code quality; large scope and multiple commits after review suggest high complexity but rapid resolution.

Summary

Add n8n Credits credential selection experiment

Adds a new experiment to let users choose between n8n credits and their own credentials, improving control over credential selection and reducing auto‑selection errors.

Health Assessment

• • Fast cycle time of under 2 hours with a single approval and no rework, indicating a smooth review process. The PR adds a moderate amount of code (~444 lines) but remains focused on a single feature, keeping risk low.

Summary

Add Group/Ungroup context menu actions

Users can now group and ungroup nodes via the context menu, improving workflow organization and reducing manual steps.

Health Assessment

• • The PR had a long cycle time of over five days and required many commits after the initial review, indicating significant rework, but the review itself was quick and no blockers were reported.

Summary

Remove assistant's embedded agent-builder toolset

Removes the assistant's embedded agent-builder toolset, temporarily disabling agent building until follow-up PRs restore this functionality.

Health Assessment

• • The PR is part of a larger effort to rework the agent builder, with multiple follow-up PRs planned. The removal of the agent builder toolset is a significant change, but the follow-up PRs will restore this functionality.

Summary

Fix AWS PrivateLink signing in HTTP Request node

Ensures HTTP requests to AWS services via PrivateLink are signed correctly, preventing signature errors and improving reliability for users leveraging AWS PrivateLink endpoints.

Health Assessment

• • The PR had a long cycle time but minimal review friction, indicating a straightforward bug fix that was eventually merged after several merge conflicts were resolved.

Summary

Track security-policy endpoints in API coverage

Adds two entries to the n8n node's API coverage manifest marking security-policy endpoints as gaps, keeping coverage in sync with the public API. No runtime changes, only documentation update.

Health Assessment

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