Large MCP response (~13.3k tokens), this can fill up context quickly

[xucian]

constantly getting really large logs from console I suppose. this is a simple project, just a few logs here and there for debugging. I can see this easily getting to 500k-1m tokens on larger projects
is there any plan to more efficiently extract logs? I assume you're getting hte full stack trace which is rarely needed, but then again 13k tokens is quite large for my 20-30 log lines

[msanatan]

It may be addressed in #287, but I'll leave this issue to track it

[dsarno]

This PR #304 fixes read_console so you can set stacktrace parameter to false correctly, should result in shorter console logs on read_console. Not sure if that's what you're referring to with the log jamming?