I find a lot of the inefficiency also comes from the model just randomly poking around and grepping all the time which is the fault of the harness. I ended up building a Prolog based MCP where I use tree-sitter to parse the code into a graph, and then the model can just ask questions like 'what are all the functions connected to this function'. So, in case you're trying to focus on what a particular endpoint is doing, you can trivially and predictably trace the whole subgraphs of calls.

https://github.com/yogthos/chiasmus

Chiasmus Looks very cool. I might have a use for it because I like to use LLM harnesses to explore code. Thanks.