Skills Reference

All 25 skills documented — arguments, integrations, examples.

🧭 Navigation · /ops · /ops:dash 📊 Daily · /ops:go · /ops:next · /ops:inbox · /ops:comms · /ops:merge 🛠️ Project & Engineering · /ops:projects · /ops:linear · /ops:triage · /ops:fires · /ops:deploy 💰 Business · /ops:revenue · /ops:ecom · /ops:marketing · /ops:voice 🤖 Automation · /ops:orchestrate · /ops:yolo 🧰 Maintenance · /ops:speedup · /ops:doctor · /ops:monitor · /ops:settings · /ops:integrate · /ops:setup · /ops:uninstall

AskUserQuestion batching (Rule 1)

Important

Every skill enforces a hard limit of ≤4 options per AskUserQuestion call (schema limit). Larger menus batch with a [More options...] bridge and paginate dynamic lists at 4 per page.

Examples in v0.8.0: setup section picker (11 items → 4+4+3), setup channel picker (7 items → 4+3), ops-comms / ops-deploy / ops-fires / ops-go / ops-inbox / ops-linear / ops-projects / ops-revenue / ops-speedup / ops-triage / ops-yolo all use "More..." bridges. ops-dash hotkey menu refactored to comply. See Plugin Rules.

Navigation

`/ops`

Business operations router. Routes to the right sub-skill, or launches the dashboard with no args.

File: skills/ops/SKILL.md Arguments: [command] [args]

Input	Routes to
(empty)	`/ops:dash`
`go`, `morning`, `briefing`	`/ops:go`
`setup`, `configure`, `init`, `install`	`/ops:setup`
`inbox`, `unread`, `messages`	`/ops:inbox`
`comms`, `send`, `whatsapp`, `email`, `slack`, `telegram`	`/ops:comms`
`fires`, `incidents`, `down`, `sentry`	`/ops:fires`
`projects`, `dashboard`, `status`	`/ops:projects`
`next`, `priority`, `what`	`/ops:next`
`triage`, `issues`	`/ops:triage`
`linear`, `sprint`, `board`	`/ops:linear`
`revenue`, `money`, `mrr`, `costs`	`/ops:revenue`
`deploy`, `ship`	`/ops:deploy`
`merge`, `prs`	`/ops:merge`
`yolo`	`/ops:yolo`
`ecom`, `shopify`, `orders`	`/ops:ecom`
`marketing`, `klaviyo`, `ads`	`/ops:marketing`
`voice`, `call`, `tts`, `transcribe`	`/ops:voice`
`orchestrate`, `autonomous`	`/ops:orchestrate`
`speedup`, `clean`, `optimize`	`/ops:speedup`
`doctor`, `health`, `fix`	`/ops:doctor`
`monitor`, `apm`, `datadog`, `newrelic`, `otel`	`/ops:monitor`
`settings`, `credentials`, `reconfigure`	`/ops:settings`
`integrate`, `partner`, `connect`	`/ops:integrate`

`/ops:dash`

Interactive pixel-art command center. Visual HQ with live status indicators and hotkey navigation.

File: skills/ops-dash/SKILL.md Arguments: [back|settings|share|faq]

Live indicators: fires · unread · PRs · GSD phases · daemon health · revenue Hotkeys: 1–9, 0, a–h, q (all within the 4-option batch rule, refactored in v0.8.0)

/ops:dash                # open dashboard
/ops:dash settings       # jump to settings
/ops:dash share          # "share your setup" social flow
/ops:dash faq            # help / FAQ / wiki links

Daily Operations

`/ops:go`

Token-efficient morning briefing. Pre-gathers data via ! shell blocks before model context loads.

File: skills/ops-go/SKILL.md Arguments: [project-alias]

Data gathered (parallel, pre-context):

flowchart LR
    G["/ops:go"] --> P[ops-gather<br/>parallel runner]
    P --> I[ops-infra → ECS]
    P --> GS[ops-git → repos]
    P --> PR[ops-prs → GitHub]
    P --> CI[ops-ci → Actions]
    P --> U[ops-unread → channels]
    P --> DC[daemon cache]
    I & GS & PR & CI & U & DC --> D[Unified dashboard]

    classDef primary fill:#6366f1,color:#fff
    classDef secondary fill:#22c55e,color:#fff
    class G,D primary
    class I,GS,PR,CI,U,DC secondary

Load time: <3s from cache (daemon pre-warm) · <10s cold

/ops:go                  # full briefing
/ops:go myapp            # scoped to one project

`/ops:next`

Priority-ordered next action. Applies the stack and routes to the right skill.

File: skills/ops-next/SKILL.md Arguments: [context]

Priority stack: fires → urgent comms → ready-to-merge PRs → Linear sprint → revenue-generating GSD work

/ops:next                  # what should I do right now?
/ops:next focus on myapp   # scoped recommendation

`/ops:inbox`

Full inbox management. Reads complete conversation threads (20+ messages), builds contact profile cards from cross-channel search, drafts replies matching your language/style.

File: skills/ops-inbox/SKILL.md Arguments: [channel: whatsapp|email|slack|telegram|all] Uses: TeamCreate("inbox-channels") · 4 parallel channel scanners

Channel	Read	Send	Archive	Requires
WhatsApp	Full inbox (7-day window)	✅	❌ (wacli limitation)	`wacli`
Email	Full inbox (max 30 threads)	✅	✅ via `gog gmail archive`	`gog` CLI
Slack	MCP + channel read	✅	❌	Slack MCP / token
Telegram	All dialogs	✅	❌	Telegram MCP (bundled)

Classification: each conversation tagged NEEDS_REPLY, WAITING, or HANDLED.

Warning

Safety rail: NEVER send without full thread understanding. The skill reads at least 20 messages and queries memories/contacts/<name>.md before drafting.

/ops:inbox               # all channels
/ops:inbox email         # Gmail only
/ops:inbox whatsapp      # WhatsApp only

`/ops:comms`

Send and read messages across any channel. Full conversation context required before any send.

File: skills/ops-comms/SKILL.md Arguments: [channel] | send [message] to [contact] | read [channel]

Note

WhatsApp health pre-flight runs automatically via the PreToolUse hook (ops-pretool-wacli-health) — the skill checks ~/.wacli/.health before any wacli call.

/ops:comms send "shipping today" to John Smith
/ops:comms send "build is green" to #deploys on Slack
/ops:comms read whatsapp
/ops:comms read telegram

`/ops:merge`

Autonomous PR merge pipeline. Dispatches parallel subagents to fix CI, resolve conflicts, address review comments, then merges.

File: skills/ops-merge/SKILL.md Arguments: [--main] [--repo org/repo] [--dry-run] Uses: TeamCreate("merge-fixers") · one agent per PR · SendMessage for cross-PR coordination

flowchart TD
    M["/ops:merge"] --> S[ops-merge-scan]
    S --> T[TeamCreate<br/>merge-fixers]
    T --> A1[fixer-repo-a]
    T --> A2[fixer-repo-b]
    T --> A3[fixer-repo-c]
    A1 & A2 & A3 --> R{Merge ready?}
    R -->|yes| OK[✅ Merge]
    R -->|CI failing| FX[Fix code, push]
    R -->|conflict| CN[Resolve + push]
    R -->|review| RC[Address comments]
    FX & CN & RC -.->|retry| R

    classDef primary fill:#6366f1,color:#fff
    classDef success fill:#22c55e,color:#fff
    classDef warn fill:#f59e0b,color:#fff
    class M,S,T primary
    class OK success
    class FX,CN,RC warn

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

Project & Engineering

`/ops:projects`

Portfolio dashboard. All registered projects with GSD phase, branch state, uncommitted files, open PRs, CI status, next suggested action.

File: skills/ops-projects/SKILL.md Arguments: [project-alias]

/ops:projects              # all projects
/ops:projects myapp        # single deep-dive

`/ops:linear`

Linear sprint board and issue management. 12 Linear MCP tools across a single surface.

/ops:linear                          # default view
/ops:linear sprint                   # current cycle
/ops:linear create "Fix login bug"   # new issue
/ops:linear HEA-42                   # specific issue
/ops:linear backlog                  # backlog review

`/ops:triage`

Cross-platform issue triage (Sentry + Linear + GitHub). Auto-resolves already-fixed issues, dispatches fix agents for active ones.

File: skills/ops-triage/SKILL.md Arguments: [project-alias|sentry|linear|github|all] Uses: TeamCreate("triage-fixers") · isolated worktrees via triage-agent

/ops:triage                # all platforms
/ops:triage sentry         # Sentry only
/ops:triage myapp          # project-scoped

`/ops:fires`

Production incidents dashboard. ECS health, Sentry errors, CI failures. Classifies by severity and offers to dispatch fix agents.

File: skills/ops-fires/SKILL.md Arguments: [project-alias|all] Uses: TeamCreate("fire-fixers") · one agent per service

Severity	Criteria
🔴 CRITICAL	Service down, DB unreachable, auth broken
🟠 HIGH	Elevated error rate, deploy stuck, main branch CI broken
🟡 MEDIUM	Non-critical service degraded, flaky tests
🟢 LOW	Warning-level, non-urgent

/ops:fires                 # all projects
/ops:fires myapp           # specific project

`/ops:deploy`

Deploy status across ECS, Vercel, GitHub Actions.

File: skills/ops-deploy/SKILL.md Arguments: [project-alias|ecs|vercel|all]

/ops:deploy                # full status
/ops:deploy vercel         # Vercel only
/ops:deploy myapp          # project-scoped

Business Intelligence

`/ops:revenue`

Revenue and costs. AWS spend via Cost Explorer, credits tracker, project revenue stages, burn rate, runway estimate.

File: skills/ops-revenue/SKILL.md Arguments: [costs|revenue|credits|runway|all]

/ops:revenue               # full dashboard
/ops:revenue costs         # AWS spend breakdown
/ops:revenue runway        # burn rate + runway

`/ops:ecom`

Shopify command center. Orders, inventory, fulfillment, analytics, store health via the Shopify Admin REST/GraphQL API.

Ships with bin/ops-shopify-create (non-interactive app scaffolding) and templates/shopify-admin-app/ (full Remix template with all admin scopes).

/ops:ecom                  # store dashboard
/ops:ecom orders           # recent orders + fulfillment
/ops:ecom inventory        # low-stock alerts
/ops:ecom analytics        # revenue, AOV, conversion
/ops:ecom setup            # configure Shopify API

`/ops:marketing`

Marketing analytics — email (Klaviyo), ads (Meta / Google), analytics (GA4), SEO (GSC), social.

/ops:marketing             # full dashboard
/ops:marketing email       # Klaviyo campaign performance
/ops:marketing ads         # Meta + Google Ads spend/ROAS
/ops:marketing seo         # SEO + GA4 organic traffic

`/ops:voice`

Voice operations — outbound calls (Bland AI), TTS (ElevenLabs), transcription (Whisper / Groq).

File: skills/ops-voice/SKILL.md Arguments: [call|tts|transcribe|setup]

/ops:voice call +15551234567 "Check in on the order"
/ops:voice tts "Your order is ready"
/ops:voice transcribe recording.mp3
/ops:voice setup           # configure API keys

Orchestration & Automation

`/ops:orchestrate`

Autonomous multi-project orchestration engine. Audits registered projects, structures work into dependency-wired tasks, dispatches parallel agents (subagents or Agent Teams), audits completions, and ships PRs.

/ops:orchestrate                       # full autonomous run (hybrid mode)
/ops:orchestrate --subagents           # fire-and-forget subagents
/ops:orchestrate --teams               # Agent Teams for coordination
/ops:orchestrate --dry-run             # preview plan without executing
/ops:orchestrate --fires-only          # only fix production incidents
/ops:orchestrate --project myapp       # single project
/ops:orchestrate --max-waves 2         # limit parallelism

`/ops:yolo`

C-suite analysis + autonomous mode. Spawns 4 parallel Opus 4.6 agents (CEO · CTO · CFO · COO). Produces unfiltered Hard Truths report. Type YOLO to hand over controls.

File: skills/ops-yolo/SKILL.md Model: claude-opus-4-6 Arguments: [YOLO|analyze|report] Uses: TeamCreate("yolo-csuite") · 4 agents in parallel

/ops:yolo                  # C-suite analysis
/ops:yolo analyze          # explicit analyze
/ops:yolo YOLO             # autonomous mode after report

Caution

YOLO autonomous mode obeys Rule 5 — destructive actions (deleting infra, rewriting git history, force-pushing, disabling CI) require explicit per-action confirmation. No silent deletions.

Maintenance

`/ops:speedup`

Cross-platform system optimizer (macOS / Linux / WSL). Disk, memory, processes, startup, network. Health score 0–100.

File: skills/ops-speedup/SKILL.md Arguments: [scan|clean|deep|auto]

/ops:speedup scan          # diagnose only
/ops:speedup clean         # quick cleanup
/ops:speedup deep          # full deep clean

`/ops:doctor`

Health check and auto-repair. Diagnoses manifest errors, broken permissions, invalid configs, stale caches, missing files — then spawns doctor-agent to fix everything automatically.

File: skills/ops-doctor/SKILL.md Arguments: [--check-only|--verbose]

/ops:doctor                # full check + auto-fix
/ops:doctor --check-only   # diagnose only
/ops:doctor --verbose      # detailed output

`/ops:monitor`

Unified APM surface. Polls Datadog, New Relic, and OpenTelemetry backends for active alerts, error traces, and entity health. Lightweight, read-only, backed by the Haiku 4.5 monitor-agent.

File: skills/ops-monitor/SKILL.md Arguments: [--watch] [--setup] [datadog|newrelic|otel|all] Spawns: monitor-agent (Haiku 4.5)

Flag	Effect
(none)	One-shot fetch across all configured backends
`--watch`	Live polling every 60s until cancelled
`--setup`	Configure Datadog / New Relic / OTEL credentials
`datadog` / `newrelic` / `otel`	Scope to a single backend

Note

monitor-agent uses Haiku 4.5 (not Sonnet) — the queries are lightweight and latency-sensitive, so a cheaper/faster model is used.

/ops:monitor                # snapshot of all backends
/ops:monitor --watch        # live poll every 60s
/ops:monitor datadog        # Datadog only
/ops:monitor --setup        # configure credentials

`/ops:settings`

Post-setup credential manager. Shows integration status (configured / missing / expired) at a glance and lets you update individual credentials without re-running the full setup wizard. Smoke-tests each integration after every update.

File: skills/ops-settings/SKILL.md Arguments: [integration]

Tip

Use /ops:settings when you need to rotate a single API key without touching any other part of your config. A full smoke test runs after each update to confirm the new credential works before persisting.

/ops:settings               # status dashboard + picker
/ops:settings stripe        # update Stripe key only
/ops:settings datadog       # update Datadog API + App keys

`/ops:integrate`

Add any SaaS API as a first-class integration. Provide the service name and the skill discovers auth patterns (via WebSearch), tests connectivity, and registers the API in your partner registry.

File: skills/ops-integrate/SKILL.md Arguments: [service-name] [--list]

Flag	Effect
(service name)	Discovers auth model, prompts for credentials, tests the connection, registers
`--list`	Lists all currently registered partners + their status

/ops:integrate stripe         # add Stripe (pre-bundled template used)
/ops:integrate plausible      # add any arbitrary SaaS
/ops:integrate --list         # show registered partners

Note

Discovered integrations are stored in preferences.json under partners[] and picked up automatically by /ops:orchestrate and /ops:settings.

`/ops:setup`

Interactive setup wizard. Installs CLIs, configures secrets, connects integrations, builds project registry. See Getting Started for the full flow.

File: skills/setup/SKILL.md Arguments: [section] Effort: high · maxTurns: 80

Tip

"Configure all" is the first option in every multi-select setup phase (Steps 1 sections, 2 CLIs, 3 channels, 4 MCPs). Hit Enter → Enter → Enter → everything installs with optimized defaults.

/ops:setup                 # full wizard
/ops:setup telegram        # just Telegram
/ops:setup doppler         # just secrets manager
/ops:setup daemon          # just daemon (install or reconcile)
/ops:setup registry        # just registry builder
/ops:setup monitor         # just monitoring (Datadog/New Relic/OTEL)

`/ops:uninstall`

Complete removal — plugin, credentials, MCPs, shell exports, cached files. Confirms each step before deletion.

File: skills/uninstall/SKILL.md Arguments: [--confirm]

Warning

Rule 5 applies. Every deletion requires explicit AskUserQuestion confirmation. The --confirm flag only skips the top-level confirmation — per-item confirmations still run.

/ops:uninstall             # guided removal
/ops:uninstall --confirm   # skip top-level confirmation only

Uh oh!

Skills Reference

Skills Reference

Contents

AskUserQuestion batching (Rule 1)

Navigation

/ops

/ops:dash

Daily Operations

/ops:go

/ops:next

/ops:inbox

/ops:comms

/ops:merge

Project & Engineering

/ops:projects

/ops:linear

/ops:triage

/ops:fires

/ops:deploy

Business Intelligence

/ops:revenue

/ops:ecom

/ops:marketing

/ops:voice

Orchestration & Automation

/ops:orchestrate

/ops:yolo

Maintenance

/ops:speedup

/ops:doctor

/ops:monitor

/ops:settings

/ops:integrate

/ops:setup

/ops:uninstall

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

claude-ops Wiki

Getting started

Reference

v2.0 — autonomy layer

v2.4 — smart home

v2.3 — competitor intel

v2.11 — telegram bot push · rotation improvements

v2.9 — voice & phone

Usage

Internals

Clone this wiki locally

`/ops`

`/ops:dash`

`/ops:go`

`/ops:next`

`/ops:inbox`

`/ops:comms`

`/ops:merge`

`/ops:projects`

`/ops:linear`

`/ops:triage`

`/ops:fires`

`/ops:deploy`

`/ops:revenue`

`/ops:ecom`

`/ops:marketing`

`/ops:voice`

`/ops:orchestrate`

`/ops:yolo`

`/ops:speedup`

`/ops:doctor`

`/ops:monitor`

`/ops:settings`

`/ops:integrate`

`/ops:setup`

`/ops:uninstall`