fix(langchain/createAgent): improve structured output handling

[christian-bromann]

This patch updates the structured output behavior for createAgent:

• it now better detects models that support structured responses
• it automatically selects structured responses if user only provides a json or Zod schema
• it adds e2e tests verifying the behavior
• it generally improves the implementation, e.g. compute decision which strategy to use after the modifyModelRequest hook of the middleware

[changeset-bot]

⚠️ No Changeset found

Latest commit: ad1caf3

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
langchainjs-docs	Ready	Preview	Comment	Sep 25, 2025 6:04am

1 Skipped Deployment

Project	Deployment	Preview	Comments	Updated (UTC)
langchainjs-api-refs	Ignored			Sep 25, 2025 6:04am

[hntrl]

I'm assuming this list is incomplete? Post V1 I'd love if we did some exploration on how to make this native to the provider packages (maintaining this list separately/ any kind of "model capability matrix" feels like asking for trouble)

(no fix needed here, just a comment)

[christian-bromann]

💯 Let's create a post v1 list