Update dependency ai to v5#244675
Merged
Samiul-TheSoccerFan merged 20 commits intoelastic:mainfrom Dec 8, 2025
Merged
Conversation
Samiul-TheSoccerFan
added
the
release_note:skip
Samiul-TheSoccerFan
added
backport:skip
Contributor
|
Dependency Review Bot Analysis 🔍 Found 1 new third-party dependencies:
Self ChecklistTo help with the review, please update the PR description to address the following points for each new third-party dependency listed above:
Thank you for providing this information! |
elastic-vault-github-plugin-prod
bot
requested a review
from a team
as a code owner
Samiul-TheSoccerFan
added
the
ci:project-deploy-elasticsearch
1 task
Contributor
|
Thanks for the PR! I've updated labels to make sure we backport this particular upgrade to 8.19/9.1/9.2/9.3 branches where we still use the old version. |
azasypkin
added
backport:all-open
Contributor
Author
|
@elasticmachine merge upstream |
Contributor
|
@Samiul-TheSoccerFan Could you complete the new dependency checklist? I noticed that there is only a single reference to one function from the new dependency. I am interested in if any alternatives were considered, but also the overall justification. Thanks! |
x-pack/solutions/search/plugins/search_playground/public/utils/stream.ts
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/public/utils/stream.ts
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/public/utils/stream.ts
Outdated
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/public/utils/stream.ts
Outdated
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/public/utils/stream.ts
Outdated
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/public/utils/stream.ts
Outdated
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/server/lib/conversational_chain.ts
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/server/lib/conversational_chain.ts
Outdated
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/server/lib/conversational_chain.ts
Outdated
Show resolved
Hide resolved
x-pack/solutions/search/plugins/search_playground/server/lib/conversational_chain.ts
Outdated
Show resolved
Hide resolved
Contributor
Author
|
@jeramysoucy : I have updated the PR description with the security checklist and tried to provide an explanation with Alternative exploration. |
Contributor
Author
|
@elasticmachine merge upstream |
TattdCodeMonkey
approved these changes
Dec 5, 2025
Contributor
Author
|
@elasticmachine merge upstream |
Contributor
💛 Build succeeded, but was flaky
Failed CI StepsMetrics [docs]Async chunks
History
|
Contributor
|
Starting backport for target branches: 8.19, 9.1, 9.2 |
1 task
Contributor
💔 All backports failed
Manual backportTo create the backport manually run: Questions ?Please refer to the Backport tool documentation |
Samiul-TheSoccerFan
added a commit
to Samiul-TheSoccerFan/kibana
that referenced
this pull request
Dec 8, 2025
## Summary - Update dependency ai to v5 - Addressed the breaking changes <img width="2552" height="1267" alt="Screenshot 2025-11-28 at 2 38 46 PM" src="https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e" /> ### Security Checklist: **Purpose** – `@ai-sdk/langchain` provides the official adapter that converts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI message stream` format. We use it in the search playground's conversational chain so the existing LangChain pipeline can feed the new createUIMessageStream/SSE infrastructure without rewriting the entire chain logic. **Justification** – AI SDK v5 moved the LangChain adapter out of the core `ai` package; the only supported way to keep LangChain models streaming into UI chunks is this package. Using the upstream adapter keeps us aligned with Vercel's protocol changes (LC stream events, chunk metadata, etc.) and avoids maintaining fragile, duplicate conversion code in Kibana. **Alternatives explored** –We looked at copying the old adapter logic, but the new UI chunk format (text, tools, reasoning, data events) changes quickly and is easy to get wrong. Reimplementing it ourselves would be brittle and hard to maintain, so we rely on the upstream package instead. **Existing dependencies** – Kibana already depends on LangChain itself and on the base `ai` SDK, but neither now exposes a LangChain to UI stream bridge. `@ai-sdk/langchain` is the upstream successor to what used to be bundled in `ai`, so it’s effectively continuing the same capability with ongoing support, making it the preferred and only viable option. ### Checklist Check the PR satisfies following conditions. Reviewers should verify this PR satisfies this list as well. - [X] [Unit or functional tests](https://www.elastic.co/guide/en/kibana/master/development-tests.html) were updated or added to match the most common scenarios --------- Co-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com> Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com> (cherry picked from commit 81c6347) # Conflicts: # .buildkite/scripts/steps/security/third_party_packages.txt # package.json # yarn.lock
This was referenced Dec 8, 2025
Contributor
Author
💚 All backports created successfully
Note: Successful backport PRs will be merged automatically after passing CI. Questions ?Please refer to the Backport tool documentation |
Samiul-TheSoccerFan
added a commit
to Samiul-TheSoccerFan/kibana
that referenced
this pull request
Dec 8, 2025
## Summary - Update dependency ai to v5 - Addressed the breaking changes <img width="2552" height="1267" alt="Screenshot 2025-11-28 at 2 38 46 PM" src="https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e" /> ### Security Checklist: **Purpose** – `@ai-sdk/langchain` provides the official adapter that converts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI message stream` format. We use it in the search playground's conversational chain so the existing LangChain pipeline can feed the new createUIMessageStream/SSE infrastructure without rewriting the entire chain logic. **Justification** – AI SDK v5 moved the LangChain adapter out of the core `ai` package; the only supported way to keep LangChain models streaming into UI chunks is this package. Using the upstream adapter keeps us aligned with Vercel's protocol changes (LC stream events, chunk metadata, etc.) and avoids maintaining fragile, duplicate conversion code in Kibana. **Alternatives explored** –We looked at copying the old adapter logic, but the new UI chunk format (text, tools, reasoning, data events) changes quickly and is easy to get wrong. Reimplementing it ourselves would be brittle and hard to maintain, so we rely on the upstream package instead. **Existing dependencies** – Kibana already depends on LangChain itself and on the base `ai` SDK, but neither now exposes a LangChain to UI stream bridge. `@ai-sdk/langchain` is the upstream successor to what used to be bundled in `ai`, so it’s effectively continuing the same capability with ongoing support, making it the preferred and only viable option. ### Checklist Check the PR satisfies following conditions. Reviewers should verify this PR satisfies this list as well. - [X] [Unit or functional tests](https://www.elastic.co/guide/en/kibana/master/development-tests.html) were updated or added to match the most common scenarios --------- Co-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com> Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com> (cherry picked from commit 81c6347) # Conflicts: # .buildkite/scripts/steps/security/third_party_packages.txt # package.json # x-pack/solutions/search/plugins/search_playground/public/utils/stream.ts # x-pack/solutions/search/plugins/search_playground/server/lib/conversational_chain.ts # yarn.lock
mbondyra
added a commit
to mbondyra/kibana
that referenced
this pull request
Dec 8, 2025
* commit '6647f813c9fa03ac0378e3d4756246e8dc4b4c76': (33 commits) [Detection Engine] Extracts Rules/Alerts/Exceptions permission to new Rules feature privileges (elastic#239634) [Agent Builder] Add Intro Tour (elastic#245551) Add datastream lifecycle support to indices metadata (elastic#245548) [Serverless] Update preconfigured connectors (elastic#245445) [Metrics][Discover] Discover to prefer line chars for time series data (elastic#244595) Update dependency @elastic/ebt to ^1.4.1 (main) (elastic#241629) [One Workflow] fix: request bodies with oneof schemas (`kibana.SetAlertsStatus`, etc) (elastic#245344) Update dependency ai to v5 (elastic#244675) Fix Discover trace waterfall behavior with duplicate spans (elastic#244984) [FSH] Migrated fs usage to kbn/fs for sample ingest (elastic#244163) Streamlang: Unskip type coercion test (elastic#245519) [Response Ops][Reporting] Fixing error in calculating delay value between retries (elastic#245431) Add TopNavMenuBeta to navigation plugin (elastic#243578) [scout] support custom servers configuration (elastic#244306) [Fleet] Run agentless background sync without dry run (elastic#245286) Fix Change Password Flaky Test (elastic#245443) Add new gap fill status for rules (elastic#242595) [Kibana Search] Move SLOs higher up in search results (elastic#245518) feat(slo): introduce find SLO instances internal route (elastic#245333) [FSH] Dropped unnecessary `fs` persistence for synthetics project code (elastic#244338) ...
Samiul-TheSoccerFan
added a commit
that referenced
this pull request
Dec 9, 2025
# Backport This will backport the following commits from `main` to `8.19`: - [Update dependency ai to v5 (#244675)](#244675) <!--- Backport version: 10.2.0 --> ### Questions ? Please refer to the [Backport tool documentation](https://github.com/sorenlouv/backport) <!--BACKPORT [{"author":{"name":"Samiul Monir","email":"150824886+Samiul-TheSoccerFan@users.noreply.github.com"},"sourceCommit":{"committedDate":"2025-12-08T17:26:35Z","message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467","branchLabelMapping":{"^v9.3.0$":"main","^v(\\d+).(\\d+).\\d+$":"$1.$2"}},"sourcePullRequest":{"labels":["release_note:skip","Team:Search","backport:all-open","ci:project-deploy-elasticsearch","v9.3.0"],"title":"Update dependency ai to v5","number":244675,"url":"https://github.com/elastic/kibana/pull/244675","mergeCommit":{"message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467"}},"sourceBranch":"main","suggestedTargetBranches":[],"targetPullRequestStates":[{"branch":"main","label":"v9.3.0","branchLabelMappingKey":"^v9.3.0$","isSourceBranch":true,"state":"MERGED","url":"https://github.com/elastic/kibana/pull/244675","number":244675,"mergeCommit":{"message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467"}}]}] BACKPORT--> --------- Co-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>
Samiul-TheSoccerFan
added a commit
that referenced
this pull request
Dec 9, 2025
# Backport This will backport the following commits from `main` to `9.1`: - [Update dependency ai to v5 (#244675)](#244675) <!--- Backport version: 10.2.0 --> ### Questions ? Please refer to the [Backport tool documentation](https://github.com/sorenlouv/backport) <!--BACKPORT [{"author":{"name":"Samiul Monir","email":"150824886+Samiul-TheSoccerFan@users.noreply.github.com"},"sourceCommit":{"committedDate":"2025-12-08T17:26:35Z","message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467","branchLabelMapping":{"^v9.3.0$":"main","^v(\\d+).(\\d+).\\d+$":"$1.$2"}},"sourcePullRequest":{"labels":["release_note:skip","Team:Search","backport:all-open","ci:project-deploy-elasticsearch","v9.3.0"],"title":"Update dependency ai to v5","number":244675,"url":"https://github.com/elastic/kibana/pull/244675","mergeCommit":{"message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467"}},"sourceBranch":"main","suggestedTargetBranches":[],"targetPullRequestStates":[{"branch":"main","label":"v9.3.0","branchLabelMappingKey":"^v9.3.0$","isSourceBranch":true,"state":"MERGED","url":"https://github.com/elastic/kibana/pull/244675","number":244675,"mergeCommit":{"message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467"}}]}] BACKPORT--> --------- Co-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>
Samiul-TheSoccerFan
added a commit
that referenced
this pull request
Dec 9, 2025
# Backport This will backport the following commits from `main` to `9.2`: - [Update dependency ai to v5 (#244675)](#244675) <!--- Backport version: 10.2.0 --> ### Questions ? Please refer to the [Backport tool documentation](https://github.com/sorenlouv/backport) <!--BACKPORT [{"author":{"name":"Samiul Monir","email":"150824886+Samiul-TheSoccerFan@users.noreply.github.com"},"sourceCommit":{"committedDate":"2025-12-08T17:26:35Z","message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467","branchLabelMapping":{"^v9.3.0$":"main","^v(\\d+).(\\d+).\\d+$":"$1.$2"}},"sourcePullRequest":{"labels":["release_note:skip","Team:Search","backport:all-open","ci:project-deploy-elasticsearch","v9.3.0"],"title":"Update dependency ai to v5","number":244675,"url":"https://github.com/elastic/kibana/pull/244675","mergeCommit":{"message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467"}},"sourceBranch":"main","suggestedTargetBranches":[],"targetPullRequestStates":[{"branch":"main","label":"v9.3.0","branchLabelMappingKey":"^v9.3.0$","isSourceBranch":true,"state":"MERGED","url":"https://github.com/elastic/kibana/pull/244675","number":244675,"mergeCommit":{"message":"Update dependency ai to v5 (#244675)\n\n## Summary\n\n- Update dependency ai to v5\n- Addressed the breaking changes\n\n<img width=\"2552\" height=\"1267\" alt=\"Screenshot 2025-11-28 at 2 38\n46 PM\"\nsrc=\"https://github.com/user-attachments/assets/44f9c824-79c5-4e7e-a0ad-216e79673f9e\"\n/>\n\n### Security Checklist:\n**Purpose** – `@ai-sdk/langchain` provides the official adapter that\nconverts LangChain's `Runnable.stream()` output into the AI SDK v5 `UI\nmessage stream` format. We use it in the search playground's\nconversational chain so the existing LangChain pipeline can feed the new\ncreateUIMessageStream/SSE infrastructure without rewriting the entire\nchain logic.\n\n**Justification** – AI SDK v5 moved the LangChain adapter out of the\ncore `ai` package; the only supported way to keep LangChain models\nstreaming into UI chunks is this package. Using the upstream adapter\nkeeps us aligned with Vercel's protocol changes (LC stream events, chunk\nmetadata, etc.) and avoids maintaining fragile, duplicate conversion\ncode in Kibana.\n\n**Alternatives explored** –We looked at copying the old adapter logic,\nbut the new UI chunk format (text, tools, reasoning, data events)\nchanges quickly and is easy to get wrong. Reimplementing it ourselves\nwould be brittle and hard to maintain, so we rely on the upstream\npackage instead.\n\n**Existing dependencies** – Kibana already depends on LangChain itself\nand on the base `ai` SDK, but neither now exposes a LangChain to UI\nstream bridge. `@ai-sdk/langchain` is the upstream successor to what\nused to be bundled in `ai`, so it’s effectively continuing the same\ncapability with ongoing support, making it the preferred and only viable\noption.\n\n\n### Checklist\n\nCheck the PR satisfies following conditions. \n\nReviewers should verify this PR satisfies this list as well.\n\n- [X] [Unit or functional\ntests](https://www.elastic.co/guide/en/kibana/master/development-tests.html)\nwere updated or added to match the most common scenarios\n\n---------\n\nCo-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>\nCo-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>","sha":"81c634708ab1405b64a5ca524b0f62d84d351467"}}]}] BACKPORT--> --------- Co-authored-by: kibanamachine <42973632+kibanamachine@users.noreply.github.com>
10 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Security Checklist:
Purpose –
@ai-sdk/langchainprovides the official adapter that converts LangChain'sRunnable.stream()output into the AI SDK v5UI message streamformat. We use it in the search playground's conversational chain so the existing LangChain pipeline can feed the new createUIMessageStream/SSE infrastructure without rewriting the entire chain logic.Justification – AI SDK v5 moved the LangChain adapter out of the core
aipackage; the only supported way to keep LangChain models streaming into UI chunks is this package. Using the upstream adapter keeps us aligned with Vercel's protocol changes (LC stream events, chunk metadata, etc.) and avoids maintaining fragile, duplicate conversion code in Kibana.Alternatives explored –We looked at copying the old adapter logic, but the new UI chunk format (text, tools, reasoning, data events) changes quickly and is easy to get wrong. Reimplementing it ourselves would be brittle and hard to maintain, so we rely on the upstream package instead.
Existing dependencies – Kibana already depends on LangChain itself and on the base
aiSDK, but neither now exposes a LangChain to UI stream bridge.@ai-sdk/langchainis the upstream successor to what used to be bundled inai, so it’s effectively continuing the same capability with ongoing support, making it the preferred and only viable option.Checklist
Check the PR satisfies following conditions.
Reviewers should verify this PR satisfies this list as well.