Run quick end-to-end smoke testing for {{COMPANY_SLUG_UPPER}} changes by tracing changed code paths through the real local stack and verifying behavior live.
Full Jira-ticket → PR pipeline (plan, implement, review, test, ship). Supports parallel and epic-mode execution. Two independent engine knobs — `provider=claude|codex` chooses the Step 7 implementer (default Claude Opus 4.7 1M, medium reasoning), and `reviewer=claude|codex` chooses the Phase 2 reviewer (default Claude). Most useful combination beyond defaults is `provider=claude reviewer=codex` (Opus implements, gpt-5.5 high reasoning hunts bugs) — different model families catch different defects.
Epic-level orchestrator for qship. Takes a Jira Epic ID, builds a topologically-sorted wave plan from its children, spawns one qship-persist.sh per ticket per wave in parallel, merges each wave into a consolidated epic branch, runs a lightweight wave-gate after each wave (migration check + targeted tests + 2 bug hunters + qbcheck, blocks only on Critical) and ONE full epic-level Phase 2 review after the last wave on the cumulative diff, and ships ONE consolidated PR per repo. Two engine knobs — `provider=claude|codex` chooses the Step 7 implementer (default Opus 4.7 1M, medium), and `reviewer=claude|codex` chooses the Phase 2 reviewer (default Claude). The diversified combo `provider=claude reviewer=codex` runs Opus implementing + gpt-5.5 high reviewing for different-family second opinion. Set `QSHIP_PER_WAVE_REVIEW=full` to restore full Phase 2 per wave, or `=none` to skip per-wave review entirely. Resumable, state-driven, idempotent.
One-time interactive configurator for the qship skill catalog. Run this immediately after installing the qship plugin (the SessionStart hook will remind you on first use). Walks through a short questionnaire (7 rounds, ~two dozen questions) plus a per-repo loop (1 repo or 100, the configurator doesn't care), renders the full skill set into your ~/.claude/skills/ tree, then verifies the install and required external plugins. Idempotent — re-runnable any time you want to change a value (added a service, new Jira project key, opted into a previously-skipped integration).
Audit API routes for the trailing-slash / 307-redirect auth bug — where a missing or extra trailing slash makes the server issue an HTTP 307 redirect and the client silently drops the Authorization header, causing intermittent 401s. Use when reviewing or debugging FastAPI/Starlette (or similar) routes, auth-token loss across redirects, or flaky authenticated endpoints.
Validate candidate bugs before reporting them — an adversarial second pass that confirms each finding is real (reproducible, correctly root-caused) and filters out false positives. Use after a bug hunt or code review to gate which findings are worth surfacing; qship runs it as the Step 10 quality gate before any bug is acted on.
Hunt for the root cause of a bug — trace symptoms back to the underlying defect and report findings, without applying fixes (hunt-only, so you stay in control of the change). Use when investigating a failure, unexpected behavior, or a flaky test and you want the true cause rather than a surface patch.
Review code changes for correctness, security, and convention adherence — the general-purpose code reviewer. Use on a diff or recent changes before committing or opening a PR to catch bugs, risky patterns, and style violations; the central review step in the qship pipeline.