I'm using the brightdata MCP, In particular binding it to a ReAct langgraph agent in python to run research tasks. I had a working implementation however since this morning I've started seeing these errors, both with the SSE and Streamable HTTP connections.

ERROR:mcp.client.sse:Error parsing server message pydantic_core._pydantic_core.ValidationError: 1 validation error for JSONRPCMessage Invalid JSON: EOF while parsing a string at line 1 column 51268 input_value='{"result":{"content":[{"...\\u000e�uF�z��'

ERROR:mcp.client.sse:Error in sse_reader httpcore.ReadError -> httpx.ReadError

Account support mentioned there is a potential problem when scraping pdfs, however I have an example where it is working where the extension is:[&formatType=PDF"]

Any info and advice would be greatly appreciated