feat(vercel): add variations for streaming/blocking, anthropic/openai and functions/ToolLoopAgent#129
Merged
constantinius merged 2 commits intomainfrom Apr 1, 2026
Conversation
… and functions/ToolLoopAgent
![sentry[bot]](https://avatars.githubusercontent.com/in/12637?s=60&v=4){kind=small}
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 2 potential issues.
Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
🔴 AI SDK Integration Test ResultsStatus: 3 regressions detected Summary
🔴 RegressionsThese tests were passing on main but are now failing: cloudflare/vercel :: Basic Embeddings Test (blocking)Error: Test execution failed: Wrangler exited with code 1 browser/langchain :: Multi-Turn LLM Test (streaming)Error: Browser test timed out (60s) nextjs/google-genai :: Multi-Turn LLM Test (blocking)Error: 3 check(s) failed: ✅ FixedThese tests were failing on main but are now passing:
🆕 New TestsPassing (71):
Failing (73): ❌ cloudflare/vercel :: Basic Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Basic Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Basic Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Basic Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Tool Call Agent Test (streaming, class, openai)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Tool Call Agent Test (streaming, class, anthropic)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Tool Call Agent Test (blocking, class, openai)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Tool Call Agent Test (blocking, class, anthropic)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Tool Error Agent Test (streaming, class, openai)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Tool Error Agent Test (streaming, class, anthropic)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Tool Error Agent Test (blocking, class, openai)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Tool Error Agent Test (blocking, class, anthropic)Error: 3 check(s) failed: ❌ cloudflare/vercel :: Vision Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Vision Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Vision Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Vision Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Long Input Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Long Input Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Long Input Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Long Input Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Conversation ID Agent Test (streaming, function, anthropic)Error: Test execution failed: Wrangler exited with code 1 ❌ cloudflare/vercel :: Conversation ID Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Conversation ID Agent Test (streaming, class, anthropic)Error: Test execution failed: Wrangler exited with code 1 ❌ cloudflare/vercel :: Conversation ID Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ cloudflare/vercel :: Conversation ID Agent Test (blocking, class, anthropic)Error: Test execution failed: Wrangler exited with code 1 ❌ nextjs/vercel :: Basic Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Basic Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ nextjs/vercel :: Basic Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Basic Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ nextjs/vercel :: Tool Call Agent Test (streaming, class, openai)Error: 3 check(s) failed: ❌ nextjs/vercel :: Tool Call Agent Test (streaming, class, anthropic)Error: 3 check(s) failed: ❌ nextjs/vercel :: Tool Call Agent Test (blocking, class, openai)Error: 3 check(s) failed: ❌ nextjs/vercel :: Tool Call Agent Test (blocking, class, anthropic)Error: 3 check(s) failed: ❌ nextjs/vercel :: Tool Error Agent Test (streaming, class, openai)Error: 3 check(s) failed: ❌ nextjs/vercel :: Tool Error Agent Test (streaming, class, anthropic)Error: 3 check(s) failed: ❌ nextjs/vercel :: Tool Error Agent Test (blocking, class, openai)Error: 3 check(s) failed: ❌ nextjs/vercel :: Tool Error Agent Test (blocking, class, anthropic)Error: 3 check(s) failed: ❌ nextjs/vercel :: Vision Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Vision Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ nextjs/vercel :: Vision Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Vision Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ nextjs/vercel :: Long Input Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Long Input Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ nextjs/vercel :: Long Input Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Long Input Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ nextjs/vercel :: Conversation ID Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Conversation ID Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ nextjs/vercel :: Conversation ID Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ nextjs/vercel :: Conversation ID Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ node/vercel :: Basic Agent Test (streaming, class, openai)Error: 2 check(s) failed: ❌ node/vercel :: Basic Agent Test (streaming, class, anthropic)Error: 2 check(s) failed: ❌ node/vercel :: Basic Agent Test (blocking, class, openai)Error: 2 check(s) failed: ❌ node/vercel :: Basic Agent Test (blocking, class, anthropic)Error: 2 check(s) failed: ❌ node/vercel :: Tool Call Agent Test (streaming, class, openai)Error: 3 check(s) failed: |
nicohrubec
reviewed
Apr 1, 2026
| {% if user_content is iterable and user_content is not string %} | ||
| const { text } = await agent.generate({ | ||
| {% if system_content %} | ||
| system: "{{ system_content }}", |
Member
There was a problem hiding this comment.
according to https://ai-sdk.dev/docs/reference/ai-sdk-core/tool-loop-agent#parameters there is no system parameter for this methods (similar for agent.stream()). system instructions need to be set using instructions in the ToolLoopAgent constructor. I think this needs to be fixed for the other templates as well
| system: "{{ system_content }}", | ||
| {% endif %} | ||
| prompt: "{{ user_content }}", | ||
| experimental_telemetry: { isEnabled: true, recordInputs: true, recordOutputs: true }, |
Member
There was a problem hiding this comment.
this also needs to be set in the constructor of the agent (same for the other templates): https://ai-sdk.dev/docs/reference/ai-sdk-core/tool-loop-agent#experimental_telemetry
Collaborator
Author
There was a problem hiding this comment.
Should be fixed now as well
| try { | ||
| {% if agentStyle == "class" %} | ||
| // ToolLoopAgent class-based approach | ||
| const agent = new ToolLoopAgent({ |
Member
There was a problem hiding this comment.
here we are missing the experimental_telemetry option
Comment on lines
+145
to
+155
| messages: [ | ||
| { | ||
| role: "user", | ||
| content: {{ renderVercelContent(user_content) }}, | ||
| }, | ||
| ], | ||
| {% if agent and agent.tools and agent.tools.length > 0 %} | ||
| tools, | ||
| stopWhen: stepCountIs(10), | ||
| {% endif %} | ||
| }); |
There was a problem hiding this comment.
Bug: The streamText calls in the function-based streaming path for the node template are missing the experimental_telemetry parameter, which prevents telemetry capture for these operations.
Severity: MEDIUM
Suggested Fix
Add the experimental_telemetry: { isEnabled: true, recordInputs: true, recordOutputs: true } parameter to the streamText calls within the function-based streaming path in src/runner/templates/agents/node/vercel/template.njk to match the implementation in the generateText calls and other templates.
Prompt for AI Agent
Review the code at the location below. A potential bug has been identified by an AI
agent.
Verify if this is a real issue. If it is, propose a fix; if not, explain why it's not
valid.
Location: src/runner/templates/agents/node/vercel/template.njk#L140-L155
Potential issue: In the node agent template, the `streamText` function calls for the
function-based streaming path are missing the `experimental_telemetry` parameter. This
omission prevents Sentry's AI SDK from capturing telemetry spans for these specific LLM
operations. This is inconsistent with the blocking `generateText` calls in the same file
and with the equivalent `streamText` calls in the Cloudflare and NextJS templates, all
of which include this parameter. The consequence is a loss of observability for
streaming operations on the node platform when using the "function" agent style, which
will cause span validation checks to fail.
nicohrubec
approved these changes
Apr 1, 2026
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
2 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.
Contributes to https://linear.app/getsentry/issue/TET-2149/add-support-for-vercel-ai-sdks-toolloopagent
Contributes to https://linear.app/getsentry/issue/TET-2164/add-variations-to-for-providers-for-the-higher-level-frameworks