Ops Go

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: no behavioural pass.

Audit 0.4.1 · source sha 5e484a8e67e5 · 29 Oct 2026Report an issue or request a re-scan

• PublisherLifecycle-Innovations-Limited GitHub organizationOwner of the repository this skill is read from
• Source folderLifecycle-Innovations-Limited/claude-ops/claude-ops/skills/ops-goThe exact folder we audited
• LicenceMITAs the repository states it
• Commit pinned for this verdict5e484a8e67e5A new audit runs when the source changes
• Repository statusNot archivedPushed today
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: OPS on-demand: This skill should be used when the user asks to "morning briefing", "/ops:ops-go", or...

---
name: ops-go
description: "OPS on-demand: This skill should be used when the user asks to \"morning briefing\", \"/ops:ops-go\", or..."
argument-hint: '[project-alias]'
allowed-tools:
  - Bash
  - Read
  - Grep
  - Glob
  - Skill
  - Agent
  - TeamCreate
  - SendMessage
  - AskUserQuestion
  - TaskCreate
  - TaskUpdate
  - TaskList
  - CronCreate
  - CronList
  - WebFetch
  # Slack — multi-workspace scan during the briefing. The flow falls back to
  # direct curl from Bash for workspaces whose token_env is not bound to the MCP.
  - mcp__claude_ai_Slack__slack_search_public_and_private
  - mcp__claude_ai_Slack__slack_read_channel
  # Notion — comments/mentions surfaced in the briefing.
  - mcp__claude_ai_Notion__notion-search
  - mcp__claude_ai_Notion__notion-fetch
  - mcp__claude_ai_Notion__notion-query-data-sources
  - mcp__claude_ai_Notion__notion-get-comments
effort: medium
maxTurns: 40
---

# OPS ► MORNING BRIEFING

Load `ops-rules` before acting. Public repo (no personal data). Outbound: one draft → one approval → one send. If `AskUserQuestion` / `Workflow` are missing, follow Rule 10 in `ops-rules` (Hermes: numbered options / two-turn Telegram card; `delegate_task`).

## Runtime Context

Before executing, load available context:

1. **Preferences**: Read `${CLAUDE_PLUGIN_DATA_DIR:-$HOME/.claude/plugins/data/ops-ops-marketplace}/preferences.json`
   - `owner` — use in the greeting header ("Good morning, [owner]")
   - `timezone` — display all timestamps in this timezone
   - `default_channels` — which channels to include in unread summary

2. **Daemon health**: Read `${CLAUDE_PLUGIN_DATA_DIR}/daemon-health.json`
   - If `action_needed` is not null → surface it before the briefing
   - Check `whatsapp-bridge` status before including WhatsApp unread counts
   - Also check bridge liveness: `lsof -i :8080 | grep LISTEN`

3. **WhatsApp pre-check**: Only include WhatsApp data if bridge is running (`lsof -i :8080 | grep LISTEN`).

## CLI/API Reference

### whatsapp-bridge (WhatsApp — mcp**whatsapp**\*)

**Bridge health** — check before any WhatsApp operation:

- Running: `lsof -i :8080 | grep LISTEN` → proceed
- Not running → `launchctl kickstart -k gui/$(id -u)/com.${USER}.whatsapp-bridge`, wait 5s

| Tool                           | Params                     | Output                                            |
| ------------------------------ | -------------------------- | ------------------------------------------------- |
| `mcp__whatsapp__list_chats`    | `{sort_by: "last_active"}` | Array of chats with jid, name, last_message_time  |
| `mcp__whatsapp__list_messages` | `{chat_jid, limit, query}` | Message array with is_from_me, content, timestamp |

### gog CLI (Gmail/Calendar)

| Command                                                             | Usage                                                      | Output                                                  |
| ------------------------------------------------------------------- | ---------------------------------------------------------- | ------------------------------------------------------- |
| `gog calendar events --all --today --json --sort start`             | Today's events across all calendars (sorted by start time) | Calendar events with `CalendarID` field for attribution |
| `gog gmail search -j --results-only --no-input --max 30 "in:inbox"` | Search inbox                                               | JSON array of threads                                   |

---

## Agent Teams support

If `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` is set, use **Agent Teams** when gathering briefing data in parallel. This enables:

- Agents share context and can coordinate mid-flight
- You can steer priorities in real-time
- Agents report progress as they complete

**Team setup** (only when flag is enabled):

```
TeamCreate("go-team")
Agent(team_name="go-team", name="infra-scanner", prompt="Check ECS health, Vercel status, and CI failures across all clusters")
Agent(team_name="go-team", name="inbox-scanner", prompt="Scan unread messages across WhatsApp, Email, Slack, Telegram, Notion")
Agent(team_name="go-team", name="pr-scanner", prompt="Find open PRs needing action — reviews, CI fixes, merge-ready")
Agent(team_name="go-team", name="sprint-scanner", prompt="Check Linear sprint progress and GSD phase state across projects")
```

If the flag is NOT set, use standard fire-and-forget subagents.

## Pre-gathered data

All data below was collected by shell scripts in <10 seconds.

### FinOps snapshot (canonical state from finops-dashboard)

Pulls current burn, anomalies, and revenue from the dashboard. Fails open
to `{}` if `FINOPS_DASHBOARD_URL` / `FINOPS_OPS_API_TOKEN` are unset — in
that case the briefing falls back to the per-source data below.

```!
${CLAUDE_PLUGIN_ROOT}/scripts/finops-bridge.sh snapshot 2>/dev/null || echo "{}"
```

**Interpreting the snapshot — two different empty states, do NOT conflate:**

1. **`{}` or unreachable** → `FINOPS_DASHBOARD_URL` / `FINOPS_OPS_API_TOKEN`
   genuinely unset (or the dashboard is down). Report "FinOps creds unset" and
   fall back to per-source AWS Cost Explorer / `Infrastructure` below.
2. **Snapshot present (non-null `generated_at`) but `services_tracked: 0` and
   all-zero `current_month_spend`/`revenue_snapshot`** → dashboard reachable +
   authenticated, but its **ingest pipeline is unpopulated**. This is NOT a
   creds problem — do NOT report "token unset". Say "FinOps dashboard reachable
   but ingest empty" and fall back to per-source data for burn/revenue.

When the snapshot has non-zero `services_tracked`, prefer it for the
burn/runway/anomaly lines. The per-source `Infrastructure` and AWS Cost
Explorer sections below are the fallback for both empty states above.

**If dashboard `current_month_spend` is ≈ $0**, dual-check with the Usage-only
helper below — credits often zero the net bill while real burn is hundreds/day.

### AWS Usage burn (RECORD_TYPE=Usage — not credit-

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

git clone https://github.com/Lifecycle-Innovations-Limited/claude-ops
git -C claude-ops checkout 5e484a8e67e5

Works in

1. 2026-10-29
   ✓SAFE B 89 / 100

   first audit · source 5e484a8e67e5

Questions about Ops Go

• github.com/Lifecycle-Innovations-Limited/claude-ops/tree/main/cl — the audited source at commit 5e484a8e67e5 and every finding aboveObserved · 29 Oct 2026

Spotted something wrong? Report a correction · Methodology