n8n@2.38.1

# [2.38.1](https://github.com/n8n-io/n8n/compare/n8n@2.37.0...n8n@2.38.1) (2026-09-01)


### Bug Fixes

* Add parameter binding for mongodb sort and projection ([#37335](https://github.com/n8n-io/n8n/issues/37335)) ([292f442](https://github.com/n8n-io/n8n/commit/292f442afa6015312e36c28d339030cdf90f1535))
* **ai-builder:** Add case for agent artifacts in AIA preview tab bar ([#37141](https://github.com/n8n-io/n8n/issues/37141)) ([ede265d](https://github.com/n8n-io/n8n/commit/ede265d46fce8cbde0e744344fb693764269f29d))
* **ai-builder:** Add several tools of the  same provider via UI ([#37212](https://github.com/n8n-io/n8n/issues/37212)) ([fd3a51c](https://github.com/n8n-io/n8n/commit/fd3a51c1149d08978c889daf15ec536b5b90d206))
* **ai-builder:** Emit tool call events for Workflow, HTTP Request and Code tools ([#37044](https://github.com/n8n-io/n8n/issues/37044)) ([7e81610](https://github.com/n8n-io/n8n/commit/7e8161043ddad5f241b15331503ddb7bb3b7e4a3))
* **ai-builder:** Fix session timeline/trace not showing on preview dock when full screen ([#36930](https://github.com/n8n-io/n8n/issues/36930)) ([80ff0f3](https://github.com/n8n-io/n8n/commit/80ff0f3ca281ce932dffa7e05df8954f41b353c5))
* **ai-builder:** Preserve managed n8n credits when building workflows ([#37198](https://github.com/n8n-io/n8n/issues/37198)) ([0fbc526](https://github.com/n8n-io/n8n/commit/0fbc52630aeb2b283da4df18517671e690510978))
* **Alibaba Cloud Node:** Use async image-generation for Wan 2.6 ([#37060](https://github.com/n8n-io/n8n/issues/37060)) ([69991a5](https://github.com/n8n-io/n8n/commit/69991a5089b4c5490277376e62ef62a706d1db85))
* **API:** Adapt binary data endpoint ([#36974](https://github.com/n8n-io/n8n/issues/36974)) ([c339069](https://github.com/n8n-io/n8n/commit/c33906903a8b82a748ae9e906dfec7d1fc431607))
* **API:** Improve config handling on instance-settings endpoints ([#36680](https://github.com/n8n-io/n8n/issues/36680)) ([a53a75d](https://github.com/n8n-io/n8n/commit/a53a75d1939664c23a8d229c5daa16e2cc35da1b))
* **API:** Resolve discover scopes from the request token grant ([#37225](https://github.com/n8n-io/n8n/issues/37225)) ([8342a46](https://github.com/n8n-io/n8n/commit/8342a466364ce4b6287be61e8f96f84e939e2d67))
* **API:** Skip projectRelations eager-load in GET /credentials ([#37138](https://github.com/n8n-io/n8n/issues/37138)) ([b3a77b0](https://github.com/n8n-io/n8n/commit/b3a77b0ba8fcb295fb6aa0d841dd43ed364204ec))
* **Chat Trigger Node:** Apply authentication to test webhook requests ([#37296](https://github.com/n8n-io/n8n/issues/37296)) ([122ee6d](https://github.com/n8n-io/n8n/commit/122ee6d5aa497ac76b17eb99ee1484c34973eaf7))
* **Chat Trigger Node:** Respect private chat setting in tests ([#37406](https://github.com/n8n-io/n8n/issues/37406)) ([0e1901f](https://github.com/n8n-io/n8n/commit/0e1901f2221a9baea3d60b48e8563abb47ab2b64))
* Copy glibc into distroless runners image as a matched pair ([#37149](https://github.com/n8n-io/n8n/issues/37149)) ([b54c9b6](https://github.com/n8n-io/n8n/commit/b54c9b6a950e7a94078cd5f9efefcb983cb1f924))
* **core:** Add a timeout to the S3 startup connection check ([#37245](https://github.com/n8n-io/n8n/issues/37245)) ([2586e21](https://github.com/n8n-io/n8n/commit/2586e21ebe3836a29939ed8d62c35db2a8308999))
* **core:** Allow a domain-restricted credential to work in its own node ([#37200](https://github.com/n8n-io/n8n/issues/37200)) ([720f397](https://github.com/n8n-io/n8n/commit/720f397947f64134c5783aff80de1265b54a7eaa))
* **core:** Apply proxy environment variables consistently across packages and processes ([#37124](https://github.com/n8n-io/n8n/issues/37124)) ([4b9dc5a](https://github.com/n8n-io/n8n/commit/4b9dc5a5ec5fab461ddf3afb03ba4fb33bb3c1b8))
* **core:** Bind OAuth token grants to their approved resource ([#37122](https://github.com/n8n-io/n8n/issues/37122)) ([380788f](https://github.com/n8n-io/n8n/commit/380788fd9a3264c83161e868026a98731d274d88))
* **core:** Detect invalid Slack Block Kit payloads in AI Assistant builds ([#37217](https://github.com/n8n-io/n8n/issues/37217)) ([b9342a5](https://github.com/n8n-io/n8n/commit/b9342a5bb8e39c7dfd9555ed372cc618542b526a))
* **core:** Fix evaluation runs failing when the Evaluation Trigger feeds the entry node ([#36975](https://github.com/n8n-io/n8n/issues/36975)) ([72283bd](https://github.com/n8n-io/n8n/commit/72283bd80241904bb22b9b3228b90ccd115cb7c2))
* **core:** Fix scripts/generate-types export path ([#32071](https://github.com/n8n-io/n8n/issues/32071)) ([106166a](https://github.com/n8n-io/n8n/commit/106166abf73d1432a7d234e176e39622b5bffefd))
* **core:** Gate browser credential creation on its own permission ([#36931](https://github.com/n8n-io/n8n/issues/36931)) ([532911c](https://github.com/n8n-io/n8n/commit/532911cd70cb63fdbc14a1333d6e44d1aa3b2bd9))
* **core:** Grant connect access for globally shared end-user credentials ([#36992](https://github.com/n8n-io/n8n/issues/36992)) ([aaf0a7e](https://github.com/n8n-io/n8n/commit/aaf0a7e7b70752477e21f6b11d315a23fa7811a6))
* **core:** Keep abandoned tool-call suspensions visible in AI Assistant thread history ([#36737](https://github.com/n8n-io/n8n/issues/36737)) ([052c30e](https://github.com/n8n-io/n8n/commit/052c30eac3d2169d739b06bd6da1444aca245b12))
* **core:** Keep AI Assistant workflow builds honest about their project ([#37177](https://github.com/n8n-io/n8n/issues/37177)) ([7fe438e](https://github.com/n8n-io/n8n/commit/7fe438e5309085d11ffeb3f7aa7e3ff09eecad37))
* **core:** Keep an existing community package directory when a fresh install fails ([#36964](https://github.com/n8n-io/n8n/issues/36964)) ([ad15ba4](https://github.com/n8n-io/n8n/commit/ad15ba4458f4c256c5c96123c4acabdaca8d1288))
* **core:** Keep the API version in the Gemini base URL for agent models ([#37239](https://github.com/n8n-io/n8n/issues/37239)) ([c2f636b](https://github.com/n8n-io/n8n/commit/c2f636bc2daea502114fe6de818506da62733252))
* **core:** Only record sub-node error metadata for nodes that get a run ([#37256](https://github.com/n8n-io/n8n/issues/37256)) ([8fcabd6](https://github.com/n8n-io/n8n/commit/8fcabd6595fa5f3cd3f53c29cd0e0e03c144e016))
* **core:** Preserve binary content in agent tool results ([#37206](https://github.com/n8n-io/n8n/issues/37206)) ([b54500d](https://github.com/n8n-io/n8n/commit/b54500d438dff056eae437efd4b69b8057024a51))
* **core:** Prevent OpenTelemetry disable from failing when OTLP collector is unreachable ([#37203](https://github.com/n8n-io/n8n/issues/37203)) ([5dd2454](https://github.com/n8n-io/n8n/commit/5dd2454f0d65b0f9e957784635a1db3d327dfb88))
* **core:** Prevent Supabase vector store node auth client timer leaks ([#37054](https://github.com/n8n-io/n8n/issues/37054)) ([ad24a48](https://github.com/n8n-io/n8n/commit/ad24a481c2fc4c33fad6bab6aed511ed80dad689))
* **core:** Re-check user tool access on agent runtime cache hits ([#37401](https://github.com/n8n-io/n8n/issues/37401)) ([6e9c3e2](https://github.com/n8n-io/n8n/commit/6e9c3e2ecfbda4cddb0e64da556d898de1cb34ff))
* **core:** Report and reconcile agent channel startup failures ([#36578](https://github.com/n8n-io/n8n/issues/36578)) ([6b0de0d](https://github.com/n8n-io/n8n/commit/6b0de0d60fa3e001c117ff4001c8c66a094c54c8))
* **core:** Report what credential setup actually selected ([#37019](https://github.com/n8n-io/n8n/issues/37019)) ([ef13ad0](https://github.com/n8n-io/n8n/commit/ef13ad0abc5692292d36b3c7ac5bd6f89af79b17))
* **core:** Respond with an error when a webhook execution ends before the Respond to Webhook node runs ([#37006](https://github.com/n8n-io/n8n/issues/37006)) ([f9814b7](https://github.com/n8n-io/n8n/commit/f9814b753f51d2fbb0be3f02add609e70d42334f))
* **core:** Restore mutating array methods on $json data in expressions ([#37172](https://github.com/n8n-io/n8n/issues/37172)) ([dcb5736](https://github.com/n8n-io/n8n/commit/dcb57362c4116075c87350c78513796aede91efa))
* **core:** Return a single not-found message when disconnecting a credential ([#37292](https://github.com/n8n-io/n8n/issues/37292)) ([bde05e0](https://github.com/n8n-io/n8n/commit/bde05e016eb3daa03e7001e50f7814b8d4804664))
* **core:** Reuse native Agent channel capabilities ([#36849](https://github.com/n8n-io/n8n/issues/36849)) ([e4edf89](https://github.com/n8n-io/n8n/commit/e4edf893c5c053b0456b93a0deeb47d0e4ae1dd6))
* **core:** Route OAuth2 token requests through environment proxies in every process ([#36996](https://github.com/n8n-io/n8n/issues/36996)) ([5266ce3](https://github.com/n8n-io/n8n/commit/5266ce3dbbb280c084833fc3fa21ffc4a62fddb5))
* **core:** Route secrets pasted in AI chat to real credential forms ([#37277](https://github.com/n8n-io/n8n/issues/37277)) ([6cc8c59](https://github.com/n8n-io/n8n/commit/6cc8c59f5ab0c9fdbe59cf8cb6f72b9e96a659ed))
* **core:** Scope active workflow IDs to the requesting user's projects ([#37326](https://github.com/n8n-io/n8n/issues/37326)) ([b557e90](https://github.com/n8n-io/n8n/commit/b557e90b0f0542370c9f1e4b99060b6fc4960ea0))
* **core:** Share one time budget across nested expression evaluations ([#37330](https://github.com/n8n-io/n8n/issues/37330)) ([99c9e79](https://github.com/n8n-io/n8n/commit/99c9e796c46b491870e751a9d7b94b3ee33a79ea))
* **core:** Split Instance Settings permissions into granular custom-role options ([#36802](https://github.com/n8n-io/n8n/issues/36802)) ([1f390cd](https://github.com/n8n-io/n8n/commit/1f390cd3593a3be71509746490df10dded13c812))
* **core:** Stage libatomic into the distroless runners image ([#37106](https://github.com/n8n-io/n8n/issues/37106)) ([ca600ad](https://github.com/n8n-io/n8n/commit/ca600ad6e67eb31e97d6a9a16925bf1f941e0ea6))
* **core:** Stop the workflow builder writing Python Code nodes that cannot run ([#37083](https://github.com/n8n-io/n8n/issues/37083)) ([16a1a78](https://github.com/n8n-io/n8n/commit/16a1a78f615202ccc5feba18c114c60da64cf7f0))
* **core:** Stop webhook routing locally before best-effort external teardown ([#37241](https://github.com/n8n-io/n8n/issues/37241)) ([cae0c8b](https://github.com/n8n-io/n8n/commit/cae0c8be00a47d5ebb74660821d60b649f89269e))
* **core:** Support writes on VM expression lazy proxies via evaluation-scoped copy-on-write ([#37262](https://github.com/n8n-io/n8n/issues/37262)) ([f77e270](https://github.com/n8n-io/n8n/commit/f77e270ffa497fa2d29e112ae0dd5dd806f766c8))
* **core:** Surface AI node failure with useful message and original cause ([#37113](https://github.com/n8n-io/n8n/issues/37113)) ([6826730](https://github.com/n8n-io/n8n/commit/68267303183fdb35a1b442de67866397a33e95e0))
* **core:** Surface the original error instead of replaying single-use bodies on OAuth2 401 retry ([#37000](https://github.com/n8n-io/n8n/issues/37000)) ([86c2a2b](https://github.com/n8n-io/n8n/commit/86c2a2b20dc04ad6bb0c9bbfb7f2509ee8f542ea))
* **core:** Teach the AI Assistant to mark polled items handled after create actions ([#37436](https://github.com/n8n-io/n8n/issues/37436)) ([78151b2](https://github.com/n8n-io/n8n/commit/78151b2205532c8db0387ec5a76a333f53859ec8))
* **core:** Thread execution id through dynamic-credential storage ([#37233](https://github.com/n8n-io/n8n/issues/37233)) ([42a782d](https://github.com/n8n-io/n8n/commit/42a782d6fb9117527d9dc2ee73533a3be09641f3))
* **core:** Unload a community package when a fresh install fails after loading it ([#37034](https://github.com/n8n-io/n8n/issues/37034)) ([78130a1](https://github.com/n8n-io/n8n/commit/78130a1a6527f3154884e8718b217333b76800e1))
* **core:** Validate the expression engine timeout and memory limit settings ([#37333](https://github.com/n8n-io/n8n/issues/37333)) ([90a84ab](https://github.com/n8n-io/n8n/commit/90a84ab28fd7b29351ac7ee9bb469ede0ba9475e))
* **core:** Verify a resolved agent model against the credential's live list ([#37240](https://github.com/n8n-io/n8n/issues/37240)) ([19582fb](https://github.com/n8n-io/n8n/commit/19582fb20d4ce5c9cf91826b66c580db10a1a8f2))
* **editor:** Attribute credential telemetry to workflow and thread ([#37027](https://github.com/n8n-io/n8n/issues/37027)) ([5fc0b75](https://github.com/n8n-io/n8n/commit/5fc0b75c4acfbdad16ec0851851368e87f5aaddc))
* **editor:** Close add-channel modal on outside click ([#37171](https://github.com/n8n-io/n8n/issues/37171)) ([c7f3ceb](https://github.com/n8n-io/n8n/commit/c7f3cebbd2fc3aea0ea8e566260c3858da45fe7e))
* **editor:** Correct community node docs link interpolation key ([#37236](https://github.com/n8n-io/n8n/issues/37236)) ([7e6cfed](https://github.com/n8n-io/n8n/commit/7e6cfed2935b98878304d0a10f84af8a2c3c5b87))
* **editor:** Distinguish session timeline errors ([#36816](https://github.com/n8n-io/n8n/issues/36816)) ([b8bc67a](https://github.com/n8n-io/n8n/commit/b8bc67a40af790b3c6cbd9ccf36ab39a3f17eeb7))
* **editor:** Gate autosave on document hydration ([#36966](https://github.com/n8n-io/n8n/issues/36966)) ([c0bc17d](https://github.com/n8n-io/n8n/commit/c0bc17d680669ee08895ce6e451b9ce04d7e8ed2))
* **editor:** Improve screen reader support for scope selector tool pills ([#37159](https://github.com/n8n-io/n8n/issues/37159)) ([e644562](https://github.com/n8n-io/n8n/commit/e64456296951c99ca226365e20b1296d36fe0100))
* **editor:** Keep workflows moved out of a folder visible in the source control push modal ([#37128](https://github.com/n8n-io/n8n/issues/37128)) ([49e28b5](https://github.com/n8n-io/n8n/commit/49e28b57ae94010c7c8458e1bb36abd65bfda7c6))
* **editor:** Render the credit banner as its own card above detached chat inputs ([#37082](https://github.com/n8n-io/n8n/issues/37082)) ([2ac3655](https://github.com/n8n-io/n8n/commit/2ac36554ffeb1db4bed51723909d2a41e125bdb5))
* **editor:** Restore parameter expression toggle layout ([#37295](https://github.com/n8n-io/n8n/issues/37295)) ([b335236](https://github.com/n8n-io/n8n/commit/b3352368ee0205ea70ecf2f526a52d55a9dd44ce))
* **editor:** Set max resize height for decision input ([#37061](https://github.com/n8n-io/n8n/issues/37061)) ([2d227dd](https://github.com/n8n-io/n8n/commit/2d227dd22a2596c9c6ed5173fb09a270a87d2f76))
* **editor:** Show an activity indicator while the AI Assistant is still working ([#37046](https://github.com/n8n-io/n8n/issues/37046)) ([5b4c65b](https://github.com/n8n-io/n8n/commit/5b4c65be393c6bf282db43a9a75d40706757c018))
* **editor:** Show error when eligible reviewers fail to load ([#37089](https://github.com/n8n-io/n8n/issues/37089)) ([45a104a](https://github.com/n8n-io/n8n/commit/45a104a15cb5015ba92b46c208a52e48aed33e38))
* **editor:** Start AI Assistant conversations in the current project ([#37051](https://github.com/n8n-io/n8n/issues/37051)) ([e5f2efc](https://github.com/n8n-io/n8n/commit/e5f2efc5568ccf286d5e6bb1094bd7b1730638a0))
* **editor:** Stop a new node from showing a deleted node's execution data ([#36915](https://github.com/n8n-io/n8n/issues/36915)) ([8bf95ac](https://github.com/n8n-io/n8n/commit/8bf95ac2b644a2a29ea33e71cb5593e33ac73902))
* **editor:** Stop autosave retries on permanent errors ([#36967](https://github.com/n8n-io/n8n/issues/36967)) ([588317e](https://github.com/n8n-io/n8n/commit/588317e0a441da2edfae714fa0e1d7fbb3865503))
* **editor:** Stop the instance AI model dropdown opening when the wizard opens ([#37215](https://github.com/n8n-io/n8n/issues/37215)) ([293e63e](https://github.com/n8n-io/n8n/commit/293e63ec2d8f3e33cf7b5e85fdeec1fc41068524))
* **editor:** Update AI credential and model selection ([#36738](https://github.com/n8n-io/n8n/issues/36738)) ([0871ad7](https://github.com/n8n-io/n8n/commit/0871ad763f333df125fdb903f384065c5f1ff7f6))
* **Evaluation Trigger Node:** Respect the row limit when reading from a data table ([#36948](https://github.com/n8n-io/n8n/issues/36948)) ([ced302a](https://github.com/n8n-io/n8n/commit/ced302ae4364a91cf10e4764abcf4d29863dab0e))
* Flag resolutions field in community node package.json vetting ([#36768](https://github.com/n8n-io/n8n/issues/36768)) ([d85c0b2](https://github.com/n8n-io/n8n/commit/d85c0b2c02478254d9b8e779bf9c15aafd22b874))
* **Form Node:** Improve error messages for invalid form definition JSON ([#37291](https://github.com/n8n-io/n8n/issues/37291)) ([d8ce73c](https://github.com/n8n-io/n8n/commit/d8ce73c1df0c2f291901f6996ad2e69b4dd772ae))
* **Gmail Trigger Node:** Do not lose queued messages when a fetch fails mid-poll ([#37263](https://github.com/n8n-io/n8n/issues/37263)) ([fe61e57](https://github.com/n8n-io/n8n/commit/fe61e57740aaf58861f4ffaf0032f57b3a8acdd5))
* **Gmail Trigger Node:** Do not miss emails when more than 100 arrive between polls ([#37261](https://github.com/n8n-io/n8n/issues/37261)) ([addc7f9](https://github.com/n8n-io/n8n/commit/addc7f94b84239e6e4b6ee8e33311a72b8bf42ea))
* **HTTP Request Node:** Keep SSL passphrases with whitespace intact ([#37020](https://github.com/n8n-io/n8n/issues/37020)) ([bcd240f](https://github.com/n8n-io/n8n/commit/bcd240f589ddfa3b363b997f4f1162fcdff9e7ab))
* **HTTP Request Node:** Show the API response when a file upload fails ([#36924](https://github.com/n8n-io/n8n/issues/36924)) ([8430516](https://github.com/n8n-io/n8n/commit/843051623f88f541517f09892e569d7662d99610))
* **Nextcloud Node:** Validate `webDavUrl` to prevent cryptic "Invalid URL" error ([#34031](https://github.com/n8n-io/n8n/issues/34031)) ([209cbfb](https://github.com/n8n-io/n8n/commit/209cbfb308f30162ed2930af074c25dc5b3c31b0))
* Preserve AWS request parameter values ([#37303](https://github.com/n8n-io/n8n/issues/37303)) ([267d98b](https://github.com/n8n-io/n8n/commit/267d98b637ab05df768ed50d63181ddd2d00f466))
* Quote and validate Postgres table and column identifiers in AI nodes ([#31370](https://github.com/n8n-io/n8n/issues/31370)) ([1fd2918](https://github.com/n8n-io/n8n/commit/1fd2918c14ac3368cbf19a5a58f1f5b6b9cf2dad))
* Require a configured model before the AI Assistant accepts a prompt ([#37226](https://github.com/n8n-io/n8n/issues/37226)) ([38e33a4](https://github.com/n8n-io/n8n/commit/38e33a49561ff9d6a2f783fa466d22d1bc6a0010))
* Restore legacy PEM headers when reformatting single-line keys ([#37088](https://github.com/n8n-io/n8n/issues/37088)) ([044a90d](https://github.com/n8n-io/n8n/commit/044a90d368228bb7b1fa80ec53eec9d81e13fa3e))
* Stop replacing error messages with "Unknown error" ([#36820](https://github.com/n8n-io/n8n/issues/36820)) ([5df7a17](https://github.com/n8n-io/n8n/commit/5df7a173a67a6fc42d16e2e0e91d2dd43e0a261d))
* Use sentence case for settings menu labels ([#36651](https://github.com/n8n-io/n8n/issues/36651)) ([7bfdc99](https://github.com/n8n-io/n8n/commit/7bfdc99049cd37de20bfb4930785e90c5599b5f5))


### Features

* Add in-widget credential status strip to @n8n/chat ([#37399](https://github.com/n8n-io/n8n/issues/37399)) ([f7f1ba3](https://github.com/n8n-io/n8n/commit/f7f1ba3be6ef5c3544fd64100395f8e6ceaa403b))
* **ai-builder:** Add an n8n credits section to the agent tools modal ([#36183](https://github.com/n8n-io/n8n/issues/36183)) ([54ac6cc](https://github.com/n8n-io/n8n/commit/54ac6cc426a8a600f6e0b6f99dba3d0f426940b6))
* **ai-builder:** Consolidate node credential resolution in agent builder to use same service as AIA ([#36903](https://github.com/n8n-io/n8n/issues/36903)) ([f41f864](https://github.com/n8n-io/n8n/commit/f41f864c0ed860d5ee82c126d602877d2f600199))
* **Confluence Node:** Add page add and remove label operations ([#37002](https://github.com/n8n-io/n8n/issues/37002)) ([ea24aa9](https://github.com/n8n-io/n8n/commit/ea24aa94ae8b2d1dcb95e1646212af45772b61a6))
* **Confluence Node:** Add Page Get Comments operation ([#37073](https://github.com/n8n-io/n8n/issues/37073)) ([843a4a8](https://github.com/n8n-io/n8n/commit/843a4a8407a5a7626344db39d0cbaac43b95c208))
* **Confluence Node:** Add Page Get Many by Label operation ([#36785](https://github.com/n8n-io/n8n/issues/36785)) ([9fbd9fb](https://github.com/n8n-io/n8n/commit/9fbd9fbc6574a47965b6d4148c7efbd477b772dc))
* **Confluence Node:** Add Space resource with Get and Get Many operations ([#36898](https://github.com/n8n-io/n8n/issues/36898)) ([58d9d39](https://github.com/n8n-io/n8n/commit/58d9d390675195ae028b18a63905d159cb872c46))
* **Confluence Node:** Page — get labels ([#36900](https://github.com/n8n-io/n8n/issues/36900)) ([9b910a1](https://github.com/n8n-io/n8n/commit/9b910a1dc97da6243a26a1d9dcfbde9d67f855ac))
* **core:** Add background execution mode for sub-agent delegation ([#37209](https://github.com/n8n-io/n8n/issues/37209)) ([4285d20](https://github.com/n8n-io/n8n/commit/4285d2063fa6dff470faa3e55c7091512e2c8568))
* **core:** Add chat-specific branding to the shared OAuth consent screen ([#37430](https://github.com/n8n-io/n8n/issues/37430)) ([e774ed2](https://github.com/n8n-io/n8n/commit/e774ed2e481ed9aaa77fb277a7905200bb8fe2e3))
* **core:** Add Moonshot, MiniMax and Qwen Cloud as model providers in Agent Builder ([#36751](https://github.com/n8n-io/n8n/issues/36751)) ([ebbdc1a](https://github.com/n8n-io/n8n/commit/ebbdc1ab72f67aec3b235e2a01aad24f2b570f3b))
* **core:** Chat trigger publish validation & workflow:execute access control ([#37231](https://github.com/n8n-io/n8n/issues/37231)) ([1f062f9](https://github.com/n8n-io/n8n/commit/1f062f9ec919901c7d847a79770cd13ec6f2c9e1))
* **core:** Establish chat session identity for credential resolution ([#37041](https://github.com/n8n-io/n8n/issues/37041)) ([c96f329](https://github.com/n8n-io/n8n/commit/c96f329fde1b4e328d3a8ec11a13ce49d83703bf))
* **core:** Let workflow verification target a specific trigger ([#37091](https://github.com/n8n-io/n8n/issues/37091)) ([6420048](https://github.com/n8n-io/n8n/commit/6420048d7adc34d9126b32f4d1effdbec81da709))
* **core:** Limit Git connections to one instance connection ([#37302](https://github.com/n8n-io/n8n/issues/37302)) ([1e4edb2](https://github.com/n8n-io/n8n/commit/1e4edb29d9e04f9e0a8c49186e141b9f20c7448c))
* **core:** Resolve sub-agents and workflow tools by run type ([#37117](https://github.com/n8n-io/n8n/issues/37117)) ([fa43bc8](https://github.com/n8n-io/n8n/commit/fa43bc88927a86c5492924f5283d6485254a9e98))
* **core:** Serve a nonce-based Content-Security-Policy on n8n's HTML pages ([#36749](https://github.com/n8n-io/n8n/issues/36749)) ([54b2b94](https://github.com/n8n-io/n8n/commit/54b2b94d566753676276315915af6dc6c9b3af4a))
* **core:** Skip live tests when n8n credits are depleted ([#36641](https://github.com/n8n-io/n8n/issues/36641)) ([a328674](https://github.com/n8n-io/n8n/commit/a3286743ac48a8e8a9b8bfdd08ab87288ee6160a))
* **core:** Split the hosted chat page into a trusted shell and a sandboxed frame ([#36947](https://github.com/n8n-io/n8n/issues/36947)) ([662a401](https://github.com/n8n-io/n8n/commit/662a401a078d9b535ba9701587bf41ee9e753e30))
* **core:** Support workflows with Wait nodes as agent tools ([#36692](https://github.com/n8n-io/n8n/issues/36692)) ([23a5f46](https://github.com/n8n-io/n8n/commit/23a5f46d3d1c3b754380d530d13af5c99c12d750))
* **Databricks Node:** Add user-delegated OAuth2 credential (authorization code + PKCE) ([#37007](https://github.com/n8n-io/n8n/issues/37007)) ([e60c431](https://github.com/n8n-io/n8n/commit/e60c43112c3f97940383ff100c4903911e00817c))
* **editor:** Add adaptive layout to ChatInput ([#36939](https://github.com/n8n-io/n8n/issues/36939)) ([435ed5f](https://github.com/n8n-io/n8n/commit/435ed5feff3b8ccdbc66bd4833ac079c2419a68c))
* **editor:** Add Contact Support link to Help menu for Cloud instances ([#37414](https://github.com/n8n-io/n8n/issues/37414)) ([e3008bd](https://github.com/n8n-io/n8n/commit/e3008bd3e5652087e0346e6c71acbebddbb355e5))
* **editor:** Add floating Markdown editor toolbar ([#37293](https://github.com/n8n-io/n8n/issues/37293)) ([5c7e006](https://github.com/n8n-io/n8n/commit/5c7e0062cb7f83f3f93e1377e97401a66a2d7d97))
* **editor:** Align workflow and agent header menus ([#37315](https://github.com/n8n-io/n8n/issues/37315)) ([848265e](https://github.com/n8n-io/n8n/commit/848265e760aac9751059e139976e75f0ae73be84))
* **editor:** Complete N8nInputNumber2 adoption and remove the legacy InputNumber ([#36598](https://github.com/n8n-io/n8n/issues/36598)) ([73ac8e4](https://github.com/n8n-io/n8n/commit/73ac8e43c8aaaffbde84b358fc40828094d7f290))
* **editor:** Rename n8n credits and n8n Connect copy to Gateway credits in the UI ([#37267](https://github.com/n8n-io/n8n/issues/37267)) ([6718392](https://github.com/n8n-io/n8n/commit/671839278608450ff97ec3b3f12f34bdc490073b))
* **editor:** Update the order of providers in the agent builder model picker ([#37143](https://github.com/n8n-io/n8n/issues/37143)) ([da6da00](https://github.com/n8n-io/n8n/commit/da6da00bd112f06347bf24afd19d7d45d8aea417))
* **Gotify Node:** Add Gotify extra notification settings ([#23915](https://github.com/n8n-io/n8n/issues/23915)) ([0aa9aa8](https://github.com/n8n-io/n8n/commit/0aa9aa8348ec309464aaf9cf61abc7814f94da71))
* **MongoDB Node:** Batch update and find-and-update writes with bulkWrite ([#37035](https://github.com/n8n-io/n8n/issues/37035)) ([5b9b332](https://github.com/n8n-io/n8n/commit/5b9b3321dcdcdf85a55e6cd4914cbb39a85e418f))
* Offer to send an anonymous failure report in the install script ([#37016](https://github.com/n8n-io/n8n/issues/37016)) ([269de76](https://github.com/n8n-io/n8n/commit/269de76c36fed60f64b5f03a43c4785147c5e2c4))
* Trigger observational memory mid-run on a token budget ([#37234](https://github.com/n8n-io/n8n/issues/37234)) ([39f49b9](https://github.com/n8n-io/n8n/commit/39f49b9a7433a916e43fb6ad2caabdd34ef8b399))


### Performance Improvements

* **core:** Acquire agent workspace sandboxes lazily, make them persistent, and share them with delegated sub-agents ([#37242](https://github.com/n8n-io/n8n/issues/37242)) ([a1b1f2e](https://github.com/n8n-io/n8n/commit/a1b1f2e9182c29e647b6da2bb2d0bfc5b7acd0cf))
* **core:** Chunk package ID lookups ([#36923](https://github.com/n8n-io/n8n/issues/36923)) ([98e5104](https://github.com/n8n-io/n8n/commit/98e51043a865b5fc7770d5870fb6edd18ded075a))
* **core:** Read the poller state row once per durable poll tick ([#36973](https://github.com/n8n-io/n8n/issues/36973)) ([be01422](https://github.com/n8n-io/n8n/commit/be014224d345e4c588c999a29267e36d90614dab))
* **core:** Reduce workflow import queries ([#36949](https://github.com/n8n-io/n8n/issues/36949)) ([bcf693e](https://github.com/n8n-io/n8n/commit/bcf693e6981f22437376564e6d1688ee6788c0fe))
* **editor:** Reduce AI gateway wallet requests by caching the credits balance ([#36725](https://github.com/n8n-io/n8n/issues/36725)) ([d62dc26](https://github.com/n8n-io/n8n/commit/d62dc265491020665752fe52ed68cd8de7f06c9b))

<!-- This is an auto-generated description by cubic. -->
<a href="https://cubic.dev/pr/n8n-io/n8n/pull/37499?utm_source=github" target="_blank" rel="noopener noreferrer" data-no-image-dialog="true"><picture><source media="(prefers-color-scheme: dark)" srcset="https://www.cubic.dev/buttons/review-in-cubic-dark.svg"><source media="(prefers-color-scheme: light)" srcset="https://www.cubic.dev/buttons/review-in-cubic-light.svg"><img alt="Review in cubic" src="https://www.cubic.dev/buttons/review-in-cubic-dark.svg"></picture></a>
<!-- End of auto-generated description by cubic. -->