Upgrade AI SDK v5 → v6 with usage null safety fixes by shrey150 · Pull Request #1694 · browserbase/stagehand

shrey150 · 2026-02-18T00:33:21Z

Summary

Upgrades ai from ^5.0.133 to ^6.0.0, @ai-sdk/provider from ^2.0.0 to ^3.0.0, and all optional AI provider packages to their latest major versions.
Migrates from LanguageModelV2 to LanguageModelV3, CoreSystemMessage/CoreUserMessage/CoreAssistantMessage to ModelMessage, and experimental_generateImage to generateImage.
Replaces deprecated generateObject/streamObject with generateText/streamText + Output.object(), with backwards-compatible shims (objectShims.ts) to preserve the existing LLMClient API surface.
Updates all agent tool toModelOutput callbacks from (result) to ({ output }) to match the v6 tool result shape.
Adds specificationVersion: "v3" to LLM logging middleware.
Fixes missing optional chaining and deprecated fallbacks on outputTokenDetails and inputTokenDetails access in both aisdk.ts and AISdkClientWrapped.ts, making token usage handling consistent across generateObject and generateText code paths.
Renames u → usage variable in generateText IIFE for consistency with the rest of the codebase.
Updates stale LanguageModelV2 comment in test file.

Based on #1689 by @dylnslck — thank you for the original upgrade work!

Test plan

pnpm install and pnpm build succeed
TypeScript typechecks pass (tsc --noEmit)
pnpm e2e:local passes — 326 passed, 2 skipped, 0 failures
Verify AISdkClient accepts LanguageModelV3 models from current provider packages without TypeScript errors

Breaking changes for external users

AISdkClient constructor now requires LanguageModelV3 instead of LanguageModelV2. Users must upgrade their @ai-sdk/* provider packages to v3+.

🤖 Generated with Claude Code

changeset-bot · 2026-02-18T00:33:32Z

🦋 Changeset detected

Latest commit: f172c5f

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 3 packages

Name	Type
@browserbasehq/stagehand	Minor
@browserbasehq/stagehand-evals	Patch
@browserbasehq/stagehand-server	Patch

Not sure what this means? Click here to learn what changesets are.

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

greptile-apps · 2026-02-18T00:41:44Z

Greptile Summary

This PR successfully migrates from AI SDK v5 to v6, upgrading all core dependencies and adapting to breaking API changes while maintaining backwards compatibility.

Key changes:

Replaces deprecated generateObject/streamObject with generateText/streamText + Output.object(), wrapped in backwards-compatible shims (objectShims.ts) to preserve the existing LLMClient API
Migrates from LanguageModelV2 to LanguageModelV3 and consolidates Core*Message types to unified ModelMessage
Updates all 8 agent tools (click, type, screenshot, etc.) to use v6 tool callback signature: toModelOutput: ({ output }) => ...
Fixes token usage access with proper null-safety: adds optional chaining for outputTokenDetails?.reasoningTokens and inputTokenDetails?.cacheReadTokens with backwards-compatible fallbacks
Adds specificationVersion: "v3" to LLM logging middleware and handles both v2 (flat numbers) and v3 (nested objects) usage shapes
Upgrades ai to ^6.0.0, @ai-sdk/provider to ^3.0.0, and all optional provider packages to their latest major versions

The migration is comprehensive and well-tested (326 e2e tests passing). The backwards-compatible shim layer ensures existing code that destructures { object } from generateObject continues to work seamlessly.

Confidence Score: 5/5

This PR is safe to merge with high confidence
The migration is systematic and comprehensive with thorough testing (326 e2e tests passing). All breaking changes are properly handled through backwards-compatible shims, token usage access is properly guarded against null values, and type migrations are complete across the entire codebase. The PR builds on previous work and includes defensive code for handling both v2 and v3 usage shapes.
No files require special attention

Important Files Changed

Filename	Overview
packages/core/lib/v3/llm/aisdk.ts	Migrates from `LanguageModelV2` to `LanguageModelV3`, replaces deprecated `generateObject` with `generateText` + `Output.object()`, adds proper null-safety for token detail access
packages/core/lib/v3/llm/objectShims.ts	New shim layer preserves backwards-compatible API surface for `generateObject`/`streamObject` callers while using v6 `generateText`/`streamText` internally
packages/core/lib/v3/external_clients/aisdk.ts	Mirrors main client migration: v2→v3 model types, `generateObject`→`generateText` + `Output.object()`, fixes token detail access patterns
packages/core/lib/v3/flowLogger.ts	Adds `specificationVersion: "v3"` to middleware, handles both v2 (flat numbers) and v3 (nested objects) usage shapes for backwards compatibility
packages/evals/lib/AISdkClientWrapped.ts	Applies same v5→v6 migration as core client with Braintrust tracing wrapper intact
packages/core/package.json	Upgrades `ai` to ^6.0.0, `@ai-sdk/provider` to ^3.0.0, and all provider packages to their latest major versions

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[AI SDK v5] --> B[AI SDK v6 Migration]
    B --> C[Update Dependencies]
    C --> D[ai package v5 to v6]
    C --> E[provider packages v2 to v3]
    C --> F[All optional providers upgraded]
    
    B --> G[Type Migrations]
    G --> H[LanguageModelV2 to V3]
    G --> I[Core*Message to ModelMessage]
    
    B --> J[API Changes]
    J --> K[generateObject deprecated]
    J --> L[streamObject deprecated]
    J --> M[Tool callback signature change]
    
    K --> N[generateText + Output.object]
    L --> O[streamText + Output.object]
    
    N --> P[objectShims.ts]
    O --> P
    P --> Q[Preserve backwards API]
    
    B --> R[Usage Shape Handling]
    R --> S[v2: flat token numbers]
    R --> T[v3: nested token objects]
    R --> U[Defensive guards added]
    
    U --> V[flowLogger middleware]
    U --> W[aisdk client]
    U --> X[external client]
    
    M --> Y[result param to output param]
    Y --> Z[All 8 agent tools updated]

_{Last reviewed commit: ac2666f}

greptile-apps

_{19 files reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

packages/core/lib/v3/llm/objectShims.ts

shrey150 · 2026-02-18T01:03:20Z

@greptileai

greptile-apps

_{21 files reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

packages/core/lib/v3/flowLogger.ts

@dylnslck

Based on #1689 by @dylnslck. Adds optional chaining and deprecated property fallbacks to token detail access across both generateObject and generateText code paths, matching the safe pattern already used in the generateObject path of aisdk.ts. Renames `u` to `usage` for consistency with the rest of the codebase. Co-Authored-By: Dylan Slack <dylnslck@users.noreply.github.com> Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

…ge fields Collapse identical prompt/messages branches in generateObjectShim and streamObjectShim into single conditions. Update v3AgentHandler to use the new v6 nested token detail fields (outputTokenDetails.reasoningTokens, inputTokenDetails.cacheReadTokens) with fallback to deprecated flat fields. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

shrey150 · 2026-02-20T16:07:40Z

@greptileai

greptile-apps

_{21 files reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

packages/core/lib/v3/flowLogger.ts

AI SDK v6's generateText/streamText return usage, finishReason, output, text, etc. as prototype getters that are lost when spread into a new object. Explicitly copy these properties so callers who destructure { object, usage, finishReason } from the shim get valid values. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

shrey150 · 2026-02-20T16:46:52Z

@greptileai

greptile-apps

_{21 files reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

cubic-dev-ai

1 issue found across 2 files (changes from recent commits).

Prompt for AI agents (all issues)


Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="packages/core/lib/v3/flowLogger.ts">

<violation number="1" location="packages/core/lib/v3/flowLogger.ts:1121">
P2: The fallback branch accessing `.total` lacks null safety. If `inputTokens` is `undefined` or `null` (e.g., a model that doesn't report usage), the `typeof` check is `false` and `.total` is accessed on `undefined`, throwing a `TypeError`. Add optional chaining and a fallback default to be consistent with the `?? 0` pattern used elsewhere in the codebase (e.g., `aisdk.ts`).</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.}

packages/core/lib/v3/flowLogger.ts

The wrapLanguageModel middleware doesn't auto-adapt v2 provider results to v3 format, so if a user passes a custom llmClient with a v2-spec model, result.usage.inputTokens is a flat number (not { total, ... }). Add typeof guards so token logging works with both spec versions. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

pirate · 2026-02-20T19:11:47Z

packages/core/lib/v3/flowLogger.ts

+            inputTokens:
+              typeof result.usage.inputTokens === "number"
+                ? result.usage.inputTokens
+                : (result.usage.inputTokens?.total ?? 0),
+            outputTokens:
+              typeof result.usage.outputTokens === "number"
+                ? result.usage.outputTokens
+                : (result.usage.outputTokens?.total ?? 0),


why would these be empty? does aisdk use different fields now?

maybe better to just do

Suggested change

inputTokens:

typeof result.usage.inputTokens === "number"

? result.usage.inputTokens

: (result.usage.inputTokens?.total ?? 0),

outputTokens:

typeof result.usage.outputTokens === "number"

? result.usage.outputTokens

: (result.usage.outputTokens?.total ?? 0),

usage: result.usage,

pirate · 2026-02-20T19:13:48Z

packages/evals/lib/AISdkClientWrapped.ts

      const result = {
-        data: objectResponse.object,
+        data: objectResponse.output,
        usage: {


Suggested change

usage: {

usage: {

...usage,

can we just return the whole obj that aisdk returns as well so users can use fields they might see in aisdk docs

pirate · 2026-02-20T19:14:03Z

packages/evals/lib/AISdkClientWrapped.ts

+        const usage = textResponse.usage;
+        return {
+          prompt_tokens: usage.inputTokens ?? 0,


Suggested change

const usage = textResponse.usage;

return {

prompt_tokens: usage.inputTokens ?? 0,

const usage = textResponse.usage;

return {

...usage,

prompt_tokens: usage.inputTokens ?? 0,

ditto here

greptile-apps bot reviewed Feb 18, 2026

View reviewed changes

packages/core/lib/v3/llm/objectShims.ts Show resolved Hide resolved

greptile-apps bot reviewed Feb 18, 2026

View reviewed changes

packages/core/lib/v3/flowLogger.ts Outdated Show resolved Hide resolved

shrey150 force-pushed the shrey/upgrade-ai-sdk-v6 branch from 8bb0449 to 9dadf0d Compare February 18, 2026 22:10

dylnslck and others added 5 commits February 20, 2026 07:47

Upgrade AI SDK and its providers

68cad20

chore: add changeset for AI SDK v6 upgrade

b3f4313

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

style: collapse Output.object() to single line

1950a24

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

shrey150 force-pushed the shrey/upgrade-ai-sdk-v6 branch from dcb29e8 to ddcea89 Compare February 20, 2026 15:48

greptile-apps bot reviewed Feb 20, 2026

View reviewed changes

packages/core/lib/v3/flowLogger.ts Outdated Show resolved Hide resolved

greptile-apps bot reviewed Feb 20, 2026

View reviewed changes

cubic-dev-ai bot reviewed Feb 20, 2026

View reviewed changes

packages/core/lib/v3/flowLogger.ts Outdated Show resolved Hide resolved

shrey150 force-pushed the shrey/upgrade-ai-sdk-v6 branch from ac2666f to f172c5f Compare February 20, 2026 16:53

pirate reviewed Feb 20, 2026

View reviewed changes

pirate approved these changes Feb 20, 2026

View reviewed changes

Conversation

shrey150 commented Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Test plan

Breaking changes for external users

Uh oh!

changeset-bot bot commented Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🦋 Changeset detected

Uh oh!

greptile-apps bot commented Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 5/5

Important Files Changed

Flowchart

Uh oh!

greptile-apps bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

shrey150 commented Feb 18, 2026

Uh oh!

greptile-apps bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

shrey150 commented Feb 20, 2026

Uh oh!

greptile-apps bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

shrey150 commented Feb 20, 2026

Uh oh!

greptile-apps bot left a comment

Choose a reason for hiding this comment

Uh oh!

cubic-dev-ai bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

pirate Feb 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

pirate Feb 20, 2026

Choose a reason for hiding this comment

Uh oh!

pirate Feb 20, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

shrey150 commented Feb 18, 2026 •

edited

Loading

changeset-bot bot commented Feb 18, 2026 •

edited

Loading

greptile-apps bot commented Feb 18, 2026 •

edited

Loading

pirate Feb 20, 2026 •

edited

Loading