Separate Before Serializing Shared State

When concurrent actors might share mutable state, first ask whether they need the same mutable object. If not, eliminate the sharing. When sharing is real, enforce serialization structurally: lockfiles, sequential phases, exclusive ownership. Instructions and conventions are not concurrency control.

Why: Concurrent writes to shared state create race conditions that are intermittent, hard to reproduce, and expensive to debug.

Pattern:

1. Identify shared mutable state (files both read and write, branches both push to, APIs both define and consume).
2. Default: eliminate the shared write target. Ask: do these actors need one canonical object, or are they publishing independent facts? Give each actor its own owned file, key, branch, or state directory, and merge only at the read/reporting boundary. Two workers writing their own lastX field into one state.json is still shared mutation. indexer-state.json + metrics-state.json is not. For parallel workers that write code, the owned unit is a git worktree (a task item with isolated: true) or a branch, and the merge happens once at the read boundary. For a stack of PRs, workers commit on their own layer and never run the commands that change the stack's topology (gh stack add, rebase, sync, push, submit, merge). One actor owns topology. A clone-wide lock makes a second writer fail loudly, but it does not make concurrent writers safe.
3. Only when one shared write target is a real invariant, serialize access structurally (lockfiles, sequential phases, single-writer actor, or atomic compare-and-swap). Treat "we need a lock" as a design smell to check, not as the default answer.