Repository navigation
fix(assessments): Fix LangGraph install, OpenRouter model names, and Mastra setup - #167
Merged
Merged
Conversation
…Mastra setup Allow @langchain/core 1.x for JS LangGraph so npm can satisfy @langchain/langgraph 1.7's ^1.3.4 peer range. Expect the provider-qualified model names that OpenRouter returns (anthropic/claude-haiku-4.5, openai/gpt-5-nano) in response model overrides. Drop the @mastra/sentry exporter. Sentry v11 instruments Mastra by default, and the second exporter re-inits Sentry and crashed the Next.js run. Co-Authored-By: Ognjen Bostjancic <ognjen.bostjancic@sentry.io>
obostjancic
marked this pull request as ready for review
November 16, 2026 10:14
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This fixes three harness problems that lower the daily score without saying anything about the SDKs.
JS LangGraph install (node + cloudflare, 0 since Nov 4). The configs pin
@langchain/coreto1.2.9, but@langchain/langgraph@1.7.4requires^1.3.4, sonpm installfails with ERESOLVE on all 4 variants. The pin is now1, the same as the llm langchain configs.@langchain/openai@1.5.10and@langchain/anthropic@1.5.8accept^1.2.9, so they still resolve.Response model overrides. Every target calls the model through OpenRouter, and OpenRouter returns names that include the provider:
anthropic/claude-haiku-4.5andopenai/gpt-5-nano. The overrides expectedclaude-haiku-4-5*andgpt-5-nano*, so 11 targets have reportedmodel.response.mismatch(major) since the reports began. The SDKs record the model the API returned, so this mismatch was never a product finding. The overrides now use the served names, the same way the Vercel configs already do.Mastra. The template registered
@mastra/sentry'sSentryExporter. Sentry v11 includesmastraIntegrationin its default tracing integrations and warns that the two must not run together, because the exporter callsSentry.init()again. On Next.js that crashed ingetDynamicSamplingContextFromSpan, so the target has always scored 1. The template now creates a plainMastraand lets the SDK instrument it.@mastra/observabilitystays installed because the SDK uses the app's copy to bootstrap the pipeline.via Ogi.
--
View Session [Sentry]