Use Cases

Use Cases

Real-world scenarios with commands and outcomes.

---

🌅 Solo founder morning routine

You start the day with 6+ tabs open — GitHub, Linear, Slack, WhatsApp, AWS console, Gmail. Context-switching costs you 20 minutes before you write a line of code.

/ops:go

Claude runs 5 shell scripts in parallel (git status, open PRs, CI failures, ECS health, unread messages), combines the results, and presents a single dashboard in under 3s (from daemon cache). One message shows: no fires, 3 WhatsApp messages, 1 PR ready to merge, sprint on track, AWS spend normal.

Then:

/ops:next

/ops:next applies the priority stack and tells you: merge the PR first (sitting for 2 days), then reply to the WhatsApp thread, then get back to the sprint.

Outcome: 60 seconds instead of 20 minutes. Full situational awareness before opening a single tab.

---

🚀 Team lead PR management

You're a team lead with 4 engineers. End of sprint, 7 open PRs across 3 repos. Some have review comments, some have CI failures, two have merge conflicts.

/ops:merge

Spawns a TeamCreate("merge-fixers") with one subagent per PR. Each reads the PR, checks CI logs, addresses review comments in code, resolves conflicts, then merges when green. Cross-PR coordination via SendMessage — when one rebases, dependents are notified.

/ops:merge --dry-run             # preview
/ops:merge --repo org/myapp      # scope to one repo
/ops:merge --main                # also sync dev → main

Outcome: 7 PRs processed in the time it would take to manually review 2. Fixable CI failures resolved automatically; judgment calls flagged for your input.

---

🚨 On-call engineer incident response

2am. PagerDuty fires. You need to know: which service is down, ECS or application-level, what Sentry is showing, is there a recent bad deploy.

/ops:fires

Pulls ECS service health for all clusters, recent Sentry errors, CI failures from last 24h, plus — thanks to the v0.8.0 infra-monitor expansion — EC2, RDS, Lambda errors, S3 public buckets, Route 53 health checks, ACM cert expiry, IAM stale keys, CloudWatch alarms, and budget overruns.

Classification: 🔴 CRITICAL · 🟠 HIGH · 🟡 MEDIUM · 🟢 LOW.

/ops:fires myapp                 # specific project

Outcome: From alert to root cause in 30 seconds. Fix agents handle straightforward issues; complex ones get a clear triage report for handoff.

---

💸 Real revenue tracking (v0.8.0)

You want actual MRR, MTD gross, and churn — not the manually-maintained numbers in registry.json.

/ops:revenue

Now pulls live data from Stripe (charges, active subs → MRR, balance, disputes, open invoices, churn) and RevenueCat (mobile subs MRR, active subscribers, churn). Combined with AWS Cost Explorer spend for net monthly + runway.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 REVENUE DASHBOARD — Apr 2026
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 MRR                  $12,450  (Stripe $10,200 + RC $2,250)
 MTD gross            $4,320
 Trailing 30d         $13,200
 Active subs          247  (vs. 241 30d ago  →  churn 2.4%)
 Open invoices        8 · $1,840
 Disputes             2 · $360
 Pending balance      $4,320   Available  $18,700
──────────────────────────────────────────────────────
 Burn rate (AWS)      $3,200/mo
 Net monthly          +$9,250  →  net positive
──────────────────────────────────────────────────────

Outcome: Actual numbers. No more "what does registry.json say again?"

---

🎯 CTO strategic review

Weekly review. You want an unfiltered picture — what's breaking, what's burning money, what's falling through the cracks — before you talk to the team.

/ops:yolo

Spawns 4 parallel Opus 4.6 agents via TeamCreate("yolo-csuite"):

• CTO — Tech debt time-bombs, services at risk, scalability limits, corners cut
• CFO — Real burn rate, waste, runway, ROI, credits expiry
• COO — Stale PRs, stalled issues, communication failures, abandoned repos
• CEO — Strategic perspective: #1 growth blocker, build-vs-buy, biggest time sinks, honest investor pitch (runs in parallel with the other three — the main /ops:yolo orchestrator synthesizes all four into the final Hard Truths report)

Outcome: CTO found your auth service has a SPOF at 3× load. CFO found $400/mo in orphaned ECS task definitions. COO found 4 PRs open >7 days with no activity.

Optional — hand over the controls:

After reading the report, type YOLO to let Claude autonomously process the inbox, merge ready PRs, fix identified fires, advance GSD phases, and deploy.

Caution

Rule 5 still applies in autonomous mode — destructive actions (delete infra, rewrite git history, force-push) always require explicit per-action confirmation. No silent deletions.

---

🌐 Remote async team coordination

Fully remote team across 3 timezones. Messages come in on Slack, WhatsApp, and occasionally email. You need to stay responsive without being always-on.

/ops:inbox

Scans all channels for the FULL inbox state (not just unread). Each conversation classified as NEEDS_REPLY, WAITING, or HANDLED. For each message needing a reply, reads at least 20 messages of thread context, queries memories/contacts/<name>.md for the contact's style, drafts a response matching their tone.

/ops:comms send "build is green, merging now" to #eng-team on Slack
/ops:comms send "can we push the call 30 mins?" to John Smith on WhatsApp

Outcome: Inbox zero across all channels in a single session. No more missing a critical Slack DM because you were in Gmail.

---

🛒 Shopify store operator

You run a Shopify store + a mobile app. You want to know today's orders, low-stock SKUs, fulfillment delays, top products, and ad ROAS in one shot.

/ops:ecom
/ops:marketing

/ops:ecom queries Shopify Admin API for orders, inventory, fulfillment status (ShipBob integration), analytics. /ops:marketing pulls Klaviyo campaigns, Meta Ads spend/ROAS, GA4 organic traffic, Search Console top queries.

Outcome: 90-second read on the entire e-commerce + marketing stack without opening Shopify admin, Klaviyo, or Facebook Business Manager.

---

☎️ Customer voice loops

A VIP customer just emailed about a delayed order. You want to call them without picking up the phone.

/ops:voice call +15551234567 "Check in on order #1247 — apologize for the delay, offer 20% off next order"

Bland AI takes the prompt, makes the call with a natural voice, and returns a transcript + summary.

Also:

/ops:voice tts "Your order has shipped" > shipping-announcement.mp3
/ops:voice transcribe interview.mp3

Outcome: High-touch customer follow-up without context-switching to another tool.

---

🤖 Full autonomous ops day

You have a heads-down coding day and want to delegate all operational work.

/ops:go                # situational awareness
/ops:next              # first action

/ops:inbox             # claude reads, drafts, archives
/ops:merge             # claude reviews, fixes CI, merges
/ops:fires             # claude triages, dispatches fix agents

/ops:yolo              # then type: YOLO

After you type YOLO, Claude enters autonomous loop mode: processes inbox, merges PRs, fixes fires, advances GSD phases, deploys. Reports back when done or when it needs a decision (Rule 5 — destructive actions always confirm).

Outcome: You wrote code for 6 hours. Claude merged 4 PRs, replied to 12 messages, resolved 2 Sentry errors, advanced 1 GSD phase, flagged 3 destructive-action decisions for you to review. End-of-day activity log waits in your inbox.

---

📟 APM war room (v1.0+)

You got a PagerDuty ping and a customer Slack message within 30 seconds of each other. You have Datadog, a New Relic account from the last migration, and an OpenTelemetry collector feeding your own Grafana. You don't want to open three tabs.

/ops:monitor

monitor-agent (Haiku 4.5) polls every configured backend in parallel for active alerts, error traces, and entity health. Returns one unified JSON surface.

/ops:monitor --watch

Live polling every 60s — one line per new alert.

/ops:monitor --backend datadog       # scope to one
/ops:monitor --setup                 # configure missing creds

Outcome: Unified APM surface across three backends in one command, Haiku-priced. When something fires, you see the correlated signal (Datadog alarm + NR error trace + OTEL span) instead of three disconnected pages.

---

🔌 Onboarding a new SaaS mid-session (v1.0+)

Your finance team just signed a contract with a payroll tool (Gusto, Rippling, whatever). You want it available to Claude immediately without hand-writing an integration.

/ops:integrate gusto

ops-integrate runs a WebSearch to discover the auth pattern (Bearer token? API key? OAuth?), shows what it found, asks you to confirm, collects the credential using the normal auto-scan → paste → deep-hunt flow, runs a health check, and registers the new partner in your local partners.json.

/ops:integrate gusto --auth bearer --url https://api.gusto.com/v1
/ops:integrate --list                # see all registered partners

From that point forward, all /ops:* skills that read partners.json can surface Gusto data the same way they surface Shopify or Klaviyo data.

Outcome: Zero-code integration onboarding. New SaaS tools are live in your briefing / inbox / automation stack in ~2 minutes, without touching plugin source.

---

⚙️ Updating credentials without re-running setup (v1.0+)

Your Stripe restricted key expired. You absolutely do not want to re-run the full /ops:setup wizard just to refresh one token.

/ops:settings --status          # show which integrations are configured / missing / expired
/ops:settings stripe            # just update Stripe, smoke-test after

ops-settings reads the current integration state, isolates the one you want to change, walks the same auto-scan → paste → smoke-test flow as the full wizard, and leaves every other integration untouched.

Outcome: 30-second credential rotation. No full-wizard re-run, no risk of clobbering unrelated settings, and you get a green/red confirmation per service.