1.98.0 (2025-06-11)

Bug Fixes

• Add a missing word to tooltip (#16198) (0cfe6ee)
• AI Agent Node: Allow using LangChain code node as a model for AI Agent (#16096) (cb3bb91)
• Anthropic Chat Model Node: Fix LmChatAnthropic node to work when both thinking is enabled and tools used (#16010) (e662998)
• Chat Memory Manager Node: Fix simplifyMessages to not overwrite consecutive messages of same type (#16168) (5015290)
• Code Node: Block selected module imports in python code (#15970) (1e92729)
• core: Correctly apply maxRows limit when fetching the dataset (#16176) (09fd5c0)
• core: Don't load task-runner on main instances when manual executions are offloaded to workers (#15986) (23ce60d)
• core: Don't swallow connection errors when fetching credentials (#16181) (9f83fcc)
• core: Ensure queue is ready when enqueueing (#16098) (25567f6)
• core: Fix HTTP proxy support in all nodes and other axios requests (#16092) (e3675bd)
• core: Prevent worker from recovering finished executions (#16094) (53b6812)
• core: Prometheus metrics can't be ingested by datadog (#15910) (d00060a)
• core: Resolves response promise for active execution on job finished in queue mode (#15643) (7cd7883)
• core: Start insights collection timer for webhook instances (#15964) (7a67dcb)
• core: Unloading a community package should also unload all its files from require.cache (#16072) (8c63ca7)
• Detect if n8n instance is staging or production for community nodes (#15918) (63cf9af)
• editor: Add right margin to AI Assistant Code Node 'Fix Error' replace code button icon (#15956) (ebff88c)
• editor: Add user role tooltip to personal settings page (#15941) (2c9c3da)
• editor: Add visual-only waitingForNext execution state for slow networks (#16143) (5deab75)
• editor: Fix regExp to allow "." in Git repository URL (SSH) (#15972) (88caa21)
• editor: Fix the issue that the content of json, html, csv, md, txt, and css files contained garbled Chinese characters when clicking the view button (#16118) (cdab4c1)
• editor: Fix the issue with icon Emojis not being centered (#16067) (b7d3945)
• editor: Hide the Github stars button when telemetry is disabled (#16108) (b92d8ad)
• editor: In Sub-workflow Conversion handle standalone accessor and accessor[0] cases correctly (#16086) (eb71c41)
• editor: Log view is missing in debug mode (#16170) (6703554)
• editor: Maintain back button when installing community nodes (#16069) (7492944)
• editor: Mark workflow as unsaved when reverting deletion of a node (#15959) (9cf7293)
• editor: Position executions filter popover so that it doesn't hide executions (#16049) (576ff4d)
• editor: Prevent Ask Assistant button overlapping other buttons (#16066) (4426249)
• editor: Refresh NDV node connections if inputs change dynamically (#16023) (02ed7b6)
• editor: Send evaluation telemetry event to Posthog (#16040) (d59b9b5)
• editor: Stop nefarious redirects during sign in (#16034) (4865d1e)
• editor: Stop nefarious URL redirection in editor middleware (#16047) (2178cfe)
• Email Trigger (IMAP) Node: Add Email UID also to email format "resolved" (#15539) (33f8fab)
• Google Sheets Node: Improve error message when row_number is null or undefined (#14560) (47538e1)
• Local File Trigger Node: Fix ignored option on Mac os (#15872) (aa40735)
• Make sure modals are listed when using a free AI credits credential in the OpenAI node (#16030) (7f7b44a)
• Microsoft SQL Node: Handle connection errors correctly with continueOnFail (#15962) (4c9198d)
• RSS Read Node: Fix issue where some feeds fail to load (#16001) (18c355d)
• Supabase Vector Store Node: Upgrade supabase SDK to address CVE-2025-48370 (#15934) (21141b4)
• Update NPM to address security vulnerability (#15968) (2e18e07)

Features

• Add Cohere reranking capability to vector stores (#16014) (8a1cabe)
• Add custom CSS to response form for HITL nodes (#15807) (3f48106)
• core: Add config to override default database ping interval and default idle connection timeout (#15764) (ac06610)
• core: Add support for partial-match execution filters (#15797) (1335af0)
• core: Allow logging JSON to stdout (#15766) (8abd556)
• core: Dynamic max page size limit for pagination dto (#16100) (5c748ca)
• core: Link 'Error Trigger' nodes to the parent execution that errored (#16016) (d64f2e5)
• core: Validate commit content for project admin role (#15687) (9607908)
• Default Data Loader Node: Add default text splitter (#15786) (40850c9)
• Default Data Loader Node: Minor copy changes (#16053) (4a6bcff)
• editor: Add "Change owner" option to editor (#15792) (5bc4e5d)
• editor: Change default node names depending on node operation and resource (#15954) (c92701c)
• editor: Enable source environment push button for project admins (#15527) (2f64809)
• editor: Improve UI for highlighted data, tags and rating in executions (#15926) (9abb333)
• editor: Make ‘Execute workflow’ a split button (#15933) (ac1a1df)
• editor: Rename debugging tab to fromai (#16017) (bc44baa)
• editor: Use node name as tool name at Vector Store retriever tool nodes (#15917) (a9f8b2d)
• n8n Node: Add missing filters (#15437) (b07c52e)
• OpenAI Chat Model Node: Update default model to gpt-4.1-mini (#15975) (0bea193)
• Structured Output Parser Node: Add auto-fix support to Strucutred Output Parser (#15915) (4f07ac3)
• Structured Output Parser Node: Add notice about $refs support in JSON schema (#15973) (47ad74d)
• Structured Output Parser Node: Mark all parameters as required for schemas generated from JSON example (#15935) (7711e4b)
• Vector Store Retriever Node: Add reranker support to retriever for QA chain (#16051) (969552a)

Performance Improvements

• core: Load source control only if licensed (#15908) (d7e94eb)
• core: Populate cache only with static webhooks (#16048) (a18822a)