crew-code-review

Deterministic discovery — candidate PRs found via gh on their own cadence, never involving the LLM; approved PRs are skipped

Eligibility holds — a quiet period stops reviews mid-rebase, a re-review cooldown gives the author room to respond to the last one

Durable DuckDB queue — candidates, positions and review history survive restarts, which is what powers Refreshed detection

Pluggable engine — codex, claude or grok, same verdict contract and same review log, so switching is a one-key config change

Steering — a PR's author can leave a short instruction for the next review, framed so it can never widen what the reviewer may do

Author scores — every reviewed PR earns its author points for a well-sized change and for net deletions, with a leaderboard

Usage floors per engine — a candidate waits when its engine's rate-limit window runs low, while the other engine's work carries on

Per-review spend recorded — token counts and API-rate cost, so a budget comes from your own data rather than a guess

Author groups as policy — ignore, comment or approve, per repo, each carrying its own engine and extra instruction

Serve with a dashboard — always-on daemon with live review logs, optionally on your tailnet; most config reloads live within ~30s

doctor checks every runtime assumption at once, and serve runs the same checks at boot