I'd open this as a discussion but since this project only has issues --
I'm extremely interested in using this for our team and I think - at the individual developer level, working via Cursor or Claude code - it solves a huge number of problems for us. We've spent months debating the best way to document and * keep up to date * the documentation for our code base (as distinct from things like design docs, testing and style requirements, etc -- stuff that we'd insert in Claude.md or as skills).
However before I go that path I want to explore how multiple teammates can use this at once. Let's say for example I host the MCP server on a VM and make it available to the engineers in our org, and I point it to main. How would this work for engineers who need to work on features? Could for example the engineers can have a local version that points to their current feature branch and simply use their favorite LLM to tie code analysis between the two together? Could/would that local version be capable of simply being a diff with the hosted version?
I may not be explaining this the best - hopefully you get my point. Essentially it's how to keep this data up to date and useful for feature development for multiple feature branches at once.
I'd open this as a discussion but since this project only has issues --
I'm extremely interested in using this for our team and I think - at the individual developer level, working via Cursor or Claude code - it solves a huge number of problems for us. We've spent months debating the best way to document and * keep up to date * the documentation for our code base (as distinct from things like design docs, testing and style requirements, etc -- stuff that we'd insert in Claude.md or as skills).
However before I go that path I want to explore how multiple teammates can use this at once. Let's say for example I host the MCP server on a VM and make it available to the engineers in our org, and I point it to main. How would this work for engineers who need to work on features? Could for example the engineers can have a local version that points to their current feature branch and simply use their favorite LLM to tie code analysis between the two together? Could/would that local version be capable of simply being a diff with the hosted version?
I may not be explaining this the best - hopefully you get my point. Essentially it's how to keep this data up to date and useful for feature development for multiple feature branches at once.