Michael Hablich
Michael Hablich
> We're using claude-agent-sdk-python and I guess they don't clean up the mcp servers that they start on exit. I think I can create an issue there. I think that's...
I think this makes sense. Similar to the extension support, we should make sure that selecting and working on other execution context does not increase the baseline token consumption of...
> thanks, I meant how can we use chrome-devtools-mcp in the built in cursor browser (without opening a separate browser instance)? Cursor 2.0 lets users check updates on the website...
I opened an issue on the Cursor issue tracker: https://github.com/cursor/cursor/issues/3888 and will close this one here, as this is related to Cursor's own browser.
Thank you, I like this too - I can see the value for this. @OrKoN what are your thoughts on this?
> I think it adds some complexity but it is doable. I wonder if there are alternatives: > > * If the client can call the tool to switch browser,...
Thanks for the FR, we are already working on something similar.
Background: We are evaluating right now how to best make LH audits and Issues available on the MCP server. @BogdanCerovac how would you expect your AI agent would make use...
I think this could be interesting for reproducibility and audit scenarios too. @OrKoN @sebastianbenz thoughts?
There is still the open question if its not the MCP client's job to handle such interactions properly.