rook-systemic-prs

Rook systemic change → small PRs

A repeatable loop for applying a systemic change (one rule applied in many places) to a github.com/rook/* repo, delivered as many small, isolated, independently mergeable PRs instead of one mega-PR. Optimized for aggressive subagent fan-out: scanning and preparation are parallelized across as many subagents as the work decomposes into, run at the harness width cap.

This skill encodes the process. The driving example throughout is dead-code elimination (the workflow it was hardened on), but the same loop applies to any sweeping change: replace a deprecated call, rename a symbol, tighten a lint, normalize imports, bump a vendored API, etc. Substitute the "find candidates" and "transform" steps for your change; the sync / exclude-open-PRs / propose-gate / per-PR-verify / conventions machinery is identical.

Core principles

1. Upstream master is the default ref. Always scan, branch, and base PRs on the current upstream master (the rook/* repo), not a stale local branch. Sync first, every run. Candidates found against stale code waste everyone's time and may already be merged.
2. Exclude work already in flight. Before proposing anything, enumerate open PRs and drop any candidate already covered by one. Re-proposing open work is the most common failure mode of a multi-session campaign.
3. Propose, then get agreement, before opening PRs. Present the candidate list and wait for the user to approve which ones to open. Do not open PRs speculatively.
4. One concern per PR. Each PR should be a single file deletion, a single dead symbol/cluster, or one mechanical transform in one area — reviewable in under a minute. Prefer whole-file deletes when an entire file is dead.
5. Fan out aggressively, up to the width cap. Decompose by directory / package / candidate file and run one subagent per unit; dispatch and width are rook-conventions references/fan-out.md. Use read-only Explore agents for scanning/auditing; use rook-maintainer:code-worker agents (shipped with this plugin) with isolation: worktree for parallel implementation across independent files.
6. Verify every PR independently. Build + vet the affected package(s) before committing. A green campaign is many green PRs, not one hopeful push.
7. Tier each agent to its work. Scan agents run a detector and report file:line — small-model work, and the whole-module cross-check in Phase 1 is what backstops it. Implementation workers apply an already-approved mechanical transform — small-model work too, backstopped by Phase 4's build/vet/gofmt and re-grep gate. Pass model: explicitly at both call sites instead of inheriting the session model. What stays on the session model is the judgment no gate can check: deciding what is safe to remove (Phase 3) and reconciling the fan-out against the authoritative pass.