Gates PRs to merge. A self-looping review-to-merge campaign: existing PRs are adopted into a run (pass PR numbers, or discover this run's labelled PRs), each PR is triaged to a review tier, and a per-PR review gauntlet (tier-dependent fresh, context-isolated SATISFIED verdicts on the whole PR diff, reviewed one at a time) plus event-driven CI monitoring gate an auto-merge. Multiple isolated runs (each keyed by a run-id, with a lease so only one agent drives each) can run concurrently in one repo. Drives its own loop through the active host's heartbeat or bounded-wait mechanism — invoke once. Campaign never writes fixes from scratch; to find issues first use gauntlet:review. Args (distinct modes): #PR... | --new #PR... | --run ID | no args
Reports findings; by default makes no changes. A two-pass adversarial code review focused on security, API consistency/symmetry, and user experience. Pass 1 is hostile (assume the worst, surface everything). Pass 2 is a neutral audit that confirms, adjusts, or refutes each finding. Use when the user asks for a hostile, skeptical, or hard review — not a friendly pass. After the report there is an opt-in handoff to gauntlet:campaign — it can open one PR per confirmed fix (implementing the fixes) and hand them to campaign to gate and merge; decline and it stays report-only, making no source/tracked-file or GitHub changes (it may write ephemeral `.gauntlet/tmp` review scratch).
Show Gauntlet's durable follow-up queue through its existing read-only table command. Use when the user asks to list or inspect Gauntlet follow-ups, deferred work candidates, or hidden and rejected follow-up entries.
Show a Gauntlet campaign ledger through its existing read-only table command. Use when the user asks for campaign status, a campaign ledger, PR state for a named or current campaign run, or hidden and merged campaign entries.
Evaluate and address GitHub Copilot PR review items for a GitHub pull request. Use when user provides a GitHub PR link and wants Copilot review comments checked, verified, fixed, committed, and summarized. Fetch review items with `gh`, verify each claim against source/tests before changing code, ask user before making subjective or constraint-driven changes, then work items one by one.