subagent-review-tier-calibration-for-overnight-pr-chainslisted

Calibrate review intensity per-PR when orchestrating a long chain (10+ PRs) of independent tasks overnight from a single implementation plan. Use when: (1) running `superpowers:subagent-driven-development` against a plan that produces many independent PRs (not stacked PRs from an issue cluster — for that shape use `overnight-multi-issue-implementation`), (2) user has authorized auto-merge on green review, (3) the user is asleep / unavailable to merge between tasks, (4) the chain spans multiple risk tiers (some PRs rewrite views/handlers, others restyle CSS only). The skill prescribes three review tiers — full two-stage / combined single-agent / bash-only verification — and a decision rubric for picking the right tier per PR to maximize overnight throughput WITHOUT silently approving real bugs. Sister to `subagent-driven-development` (specializes its review step for throughput-tuned chains) and `overnight-multi-issue-implementation` (different problem shape — issues vs. plan tasks; stacked vs. independent). Sy
wan-huiyan/overnight-workflows · ★ 1 · AI & Automation · score 71

Install: claude install-skill wan-huiyan/overnight-workflows

# Subagent Review-Tier Calibration for Overnight PR Chains ## Problem `superpowers:subagent-driven-development` prescribes a strict two-stage review per task: spec compliance reviewer first, then code-quality reviewer. For a single feature or a 3-task plan, that's correct — quality dominates throughput. For an **overnight chain of 10+ independent PRs** from one plan, blindly following two-stage means 20+ review subagent dispatches with sequential dependencies (next task gates on previous merge). Even when individual subagents take 5 minutes, the chain wall time balloons to 3-4 hours of review alone, and the orchestrator context burns through skill-list reminders. The fix is **per-PR tier selection**, not blanket two-stage: - **Tier 1 (two-stage, full)** — for PRs touching data contracts, handler architecture, or shared schemas - **Tier 2 (combined single-agent)** — for moderate-risk PRs (new template + view + tests, well-scoped) - **Tier 3 (bash-only verification)** — for low-risk PRs (pure visual restyle, no flow change, all tests green in implementer report) Calibrate per-PR, not per-chain. Some chains run Tier 1 throughout (high-risk migration); others can use Tier 3 for the bulk and reserve Tier 1 for the 2-3 high-risk PRs. ## Context / Trigger Conditions Apply this calibration when **all** of these hold: 1. You're orchestrating `superpowers:subagent-driven-development` against a plan with 10+ tasks 2. The user has authorized auto-merge on green review 3. Tasks ar