codemcp

I noticed that Claude completely changes its character when it starts using MCP. Can’t have a conversation, instructions in my memory are ignored including code styles. And it just doesn’t know how to stop. When I tell it to run tests, it will continue running and fixing them until they pass. Which is what you want often, but it can also uncontrollably spiral into its own hallucinations. https://github.com/ezyang/codemcp Is really nice because it’s a lot cheaper, but this kind of stuff and the lack of caching is pushing me towards multiple chained agents calling the API directly.

Try SDGLBL/mcp-claude-code or ezyang/codemcp. They are both very good (better then Claude Code in my opinion). Using aider is another option. Or Assistant in Zed. - diferent approaches but both very effective. https://github.com/SDGLBL/mcp-claude-code https://github.com/ezyang/codemcp

I’ve been using codemcp (https://github.com/ezyang/codemcp) to get “most” of the functionality of Claude code (I believe it uses prompts extracted from Claude Code), but using my existing pro plan. It’s less autonomous, since it’s based on the Claude chat interface, and you need to write “continue” every so often, but it’s nice to save the $$

I'm going to try out yours (https://github.com/ezyang/codemcp). I already pay for Cursor, but I'm curious. EDIT: I really like the way that each change generates a commit, and that all commits in a single session are squashed into one commit, whilst preserving the hash of each individual change. EDIT2: I also like you only have one function (each command being a 'subtool'), which means Claude Desktop asks for permission only once per session.

Most of the problem is installing the MCP servers, which is more annoying on Windows. https://github.com/ezyang/codemcp#getting-started has instructions that I've personally tested for installation on Windows, which might help you out some.

You can try (self promo) https://github.com/ezyang/codemcp . https://github.com/rusiaaman/wcgw is also quite popular, although they allow unrestricted shell access (that's why it's named wcgw lol).

Hi all! I normally work on the PyTorch project but I've been on baby leave for the past month, so I've been playing around with AI as a user rather than a framework implementor. I really liked the agent experience with Claude Code, but I couldn't really justify spending so many dollars on API costs for random side projects. I already pay for a Claude Pro subscription though, and it turns out you can simulate many of Claude Code's features with an MCP. If you have a Pro subscription, check this out! I think it really captures the Claude Code experience quite well, without forcing you to pay for API tokens.