[glamp]

Hey Boris,

I couldn't agree more. And using Plan mode was a major breakthrough for me. Speaking of Plan Mode...

I was previously using it repeatedly in sessions (and was getting great results). The most recent major release introduced this bug where it keeps referring back to the first plan you made in a session even when you're planning something else (https://github.com/anthropics/claude-code/issues/12505).

I find this bug incredibly confusing. Am I using Plan Mode in a really strange way? Because for me this is a showstopper bug–my core workflow is broken. I assume I'm using Claude Code abnormally otherwise this bug would be a bigger issue.

[fourthark]

Yes as lostdog says, it’s a new feature that writes plans in plan mode to ~/.claude/plans. And it thinks it needs to continue the same plan that it started.

So you either need to be very explicit about starting a NEW plan if you want to do more than one plan in a session, or close and start a new session between plans.

Hopefully this new feature will get less buggy. Previously the plan was only in context and not written to disk.

[manmal]

Why don’t you reset context when working on something else?

[mrieck]

It’s additional features that are related.

For example making a computer use agent… Made the plan, implementation was good, now I want to add a new tool for the agent, but I want to discuss best way to implement this tool first.

Clearing context means Claude forgets everything about what was just built.

Asking to discuss this new tool in plan mode makes Claude rewrite entire spec for some reason.

As workaround, I tell Claude “looks good, delete the plan” before doing anything. I liked the old way where once you exit plan mode the plan is done, and next plan mode is new plan with existing context.

[manmal]

I get where you're coming from. But you'll likely get better results by starting fresh and letting it read key files or only just a summary of the project goals/spec. And then implement the next feature building up on the previous one. It's unlikely you'll need all the underlying code of the foundation in context to implement something that builds up on it - especially if interfaces are clean. Models still get dumber the more context is loaded, and the usable window isn't all that big, so starting fresh gives best results usually. I try to avoid compaction in any way possible, and I rarely continue the session after compaction, for that reason.

[lostdog]

Yes, I've also been confused by things like this. Claude code is sometimes saving plans to ~/.claude/plans under animal names. But it's not really surface where the plan goes, not what the expected way to refer back to them is?

[theshrike79]

It's a cache pretty much. Before it wrote them to the project directory by default, which is really annoying.

Now it has a file it can refer to (call it "memory" to be fancy) without having to keep everything in context. The plan in the file survives over autocompact a lot better and it can just copy it to the project directory without rewriting it from memory.