91 points by azhenley about 12 hours ago | 11 comments | View on ycombinator
ptmkenny about 8 hours ago |
maxbeech about 9 hours ago |
the abstraction is right - isolated worktree, scoped task, commit only what belongs. the failure is enforcement. git worktrees don't prevent a process from running `cd ../main-repo`. that requires something external imposing the boundary, not relying on the agent to respect it.
what you've built (the 8:47 sweep) is a narrow-scope autonomous job: well-defined inputs, deterministic outputs, bounded time. these work well because the scope is clear enough that deviation is obvious. the harder category is "fix any failing tests" - that task requires judgment about what's in scope, and judgment is exactly where worktree escapes happen.
i've been working on tooling for scheduling this kind of claude work (openhelm.ai) and the isolation problem is front and center. separate working directories per run, no write access to the main repo unless that's the explicit task. your experience here is exactly the failure mode that design is trying to prevent.
sneg55 about 5 hours ago |
The iOS section is the real story though. The TCC.db dance and AppleScript keyboard limitations feel like archaeological excavation.
devmor about 10 hours ago |
When struggling with failing to reverse engineer mobile apps for smart home devices, I’ve considered trying to set something like this up for a single device.
rolifromhermes about 3 hours ago |
robutsume about 9 hours ago |
johnwhitman about 7 hours ago |
leontloveless about 9 hours ago |