feat: ./setup --enforcement for plan enforcement#409
Open
Yann-BILIEN wants to merge 2 commits intogarrytan:mainfrom
Open
feat: ./setup --enforcement for plan enforcement#409Yann-BILIEN wants to merge 2 commits intogarrytan:mainfrom
Yann-BILIEN wants to merge 2 commits intogarrytan:mainfrom
Conversation
Adds an optional --enforcement flag to ./setup that runs npx rippletide-code@latest connect, wiring Rippletide's open source Claude Code hooks for plan decision enforcement. Without --enforcement, gstack works exactly as before, no change in behavior, no new dependency.
Yann-BILIEN
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Sharing how I use gstack in case it's useful to others.
I run multiple Claude Code tabs in parallel, typically 5 or more, each working on a different part of a feature. The plan review skills (/plan-eng-review, /plan-ceo-review, /plan-design-review) are what make that possible: they force real decisions before the agent starts touching code.
The issue I kept running into is that those decisions don't carry across tabs or sessions. Tab 2 doesn't know what tab 1 decided. The agent drifts, contradicts a constraint that was already agreed, or re-opens a question that was already closed. I'd end up checking in constantly, which defeats the point of running in parallel.
What fixed it for me was adding Rippletide (open source) on top. It hooks into Claude Code and injects prior plan decisions before every prompt, and validates writes against them before files are touched. Once that's in place, I can actually leave the tabs running. The enforcement happens at the hook level so I don't have to be the one enforcing it.
This PR just adds
--enforcementto./setupso that runningnpx rippletide-code@latest connectis one step instead of something you have to look up separately:Completely optional, without it gstack works exactly as before. If you try it, check
CLAUDE.mdafterward since Rippletide generates its own and the## gstacksection needs to still be there.