What's the point of Serena regex tools and how to disable them?

[GGGSPB]

The regex find and replace tools seem to be slower and more cumbersome than the built in search and replace grep tools Claude Code has per default. It constantly breaks syntax, doesn't seem to read files before writing to them, and goes off in loops trying to use regex to find the syntax error it caused, to then break something else in the process of fixing it.

Why would I want to use the Serena regex tools over the CC ones and how do I disable them persistently?

I've been using Serena for a few weeks now and really happy with it overall, the regex functionality is the main thing that feels disjointed and awkward.

[MischaPanch]

You can disable tools through the config, see here
https://github.com/oraios/serena?tab=readme-ov-file#configuration

There are multiple reasons for the existence of this tool:

1. Serena is not only for Claude code
2. Even within Claude code, the editing operations used to be full string replacement, which is not token efficient. We don't know whether they still are based on full string replacement.

We will consider disabling this tool in the ide-assistant context in the future though, seems like editing in many coding assistants did get more efficient in the last months

[GGGSPB]

Thanks I will take a look at it and thank you for Serena MCP!