langchainjs · Issues· 568 open

RunnableWithFallbacks.batch({ returnExceptions: true }) throws instead of returning errors

Cannot create `ChatGoogle` LLM with `GOOGLE_APPLICATION_CREDENTIALS`

Let the HITL `when` predicate read tool metadata, and add an `interruptOn` catch-all

Background callbacks run in another request's AsyncLocalStorage context when the callback queue is busy

ChatOpenAI drops explicit reasoning for gpt-6-astra despite reasoning-enabled model profile

Community integration: langchain-kudosity — SMS, MMS and RCS messaging tools

createAgent().streamEvents() inside a LangGraph node delivers config callbacks twice when LangSmith tracing is enabled

createToolCallTransformer: a tool that throws crashes the process with an unhandled rejection under streamEvents v3

streamUsage default diverges from Python: setting a custom baseURL disables it in langchain-openai but not in @langchain/openai

DOC: ChatOpenAI endpoint resolution — OPENAI_API_BASE takes precedence over OPENAI_BASE_URL, documented in Python but not in JS

Would you consider adding MarkdownHeaderTextSplitter to Javascript LangChain?

google: Can't capture request and response when using streamEvents()

EventStreamCallbackHandler warns "Run ID … not found in run map" for every run when tracing is enabled and a tool invokes a nested graph (duplicate handler delivery not covered by #11190)

ChatAnthropic `ContentBlock.Tools.ServerToolCallResult` missing critical details and metadata for Web Search

bug(core): RunnableBranch._invoke falls through to default on falsy branch output (0/""/false)