Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/vscarpenter/gsd-task-manager/night-shiftgit clone --depth 1 https://github.com/vscarpenter/gsd-task-managerWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.01067 |
| Opus 5 | $0.00000 | $0.00534 |
| Sonnet 5 | $0.00000 | $0.00213 |
| Haiku 4.5 | $0.00000 | $0.00107 |
Grade A, and why
night-shift scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Night shift — operating spec
The night shift is a nightly, unattended local Claude Code routine (cycle D — the maintenance loop). At 20:00 it triages failing checks on the agent fleet's own open PRs, clearing mechanical failures before they cost you attention. It never merges. It is launched by scripts/triage-run.sh; its per-run instructions live in .claude/commands/triage-prs.md; this file is the durable operating spec both reference.
Issue/PR operations follow docs/agents/issue-tracker.md. Triage-label vocabulary is in docs/agents/triage-labels.md. Definition of done for any fix is coding-standards.md.
The nightly loop
20:00 launchd → triage-run.sh
triage:paused set? → post one line, exit (kill switch)
find open claude/* PRs with a failing check → none? exit
else: worktree off origin/main → claude -p "/triage-prs"
→ for each failing claude/* PR (oldest first, ≤3 fix PRs):
reproduce → classify → fix+verify → fix PR | escalate | skip
→ post self-audit report to the Night Shift Control issue
Scope is claude/* PRs only — the fleet's own output. It never touches hand-authored branches.
Auto-fix / escalate / skip policy
| Failing check | Action |
|---|---|
| Lint / formatting drift | Fix — eslint . --fix, prettier --write |
Stale bun.lock |
Fix — bun install |
Branch behind main (fails only for being out of date) |
Fix — merge origin/main (trivial conflicts only) |
| Simple typecheck / import error (unused/missing import, obvious annotation) | Fix — targeted, no logic changes |
| Logic test failure | Escalate (ready-for-human + reason) |
| Any security-scan failure | Escalate — never patch around |
| Ambiguous type error / non-trivial conflict / a fix that didn't verify | Escalate |
Can't reproduce / flaky / external / already ready-for-human |
Skip + log |
Every fix is delivered as a fix PR on a fresh claude/fix-<branch>-<runid> branch targeting the failing branch, so it re-enters review + CI. The night shift never merges.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 58 lines · 0 tokens per session scan A 376957352995
night-shift is an agent published in the GitHub repository vscarpenter/gsd-task-manager (24 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,067 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
mcp-tool-developer
MCP tool developer agent for building S&P 500 MCP server tools.
ot-architect
Usar para decisiones de arquitectura de OpenTicket — límites del core framework-free (core/), diseño del AgentCommerceAdapter, nuevos rails, extracción del paquete OSS, o cualquier cambio que cruce core/ ↔ lib/ ↔ app/. Invocar ANTES de escribir código que toque core/adapter o ports.ts.
ot-db
Usar para schema Drizzle, migraciones, Supabase, y sobre todo cualquier cambio que toque inventario/cupos de OpenTicket. Invocar antes de modificar db/schema.ts o db/store.ts, y para operar la DB (push, seed, advisors de Supabase).
ot-devops
Usar para CI/CD, deploy a Vercel, variables de entorno, GitHub Actions, y configuración de Supabase/Stripe/Resend en producción para OpenTicket. Invocar para el primer deploy, para armar el workflow de CI, o cuando falle un build/deploy.
ot-frontend
Usar para la web humana de OpenTicket — landing dev-native estética CLI, ticker en vivo, página pública de evento, dashboard organizador, página de orden. Invocar para cualquier cambio de UI/UX en app/.
ot-payments
Usar para todo lo que toque Stripe en OpenTicket — Checkout Sessions, webhooks, idempotencia, reembolsos, expiración de reservas, o bugs de doble emisión/doble cobro. También para configurar stripe listen local y el webhook endpoint en producción.