Skip to content
Development
Command

/portfolio-sweep

Sweep every connected security tool across all clients/tenants, normalize findings, and report the top most urgent items portfolio-wide

From plugin
msp-claude-plugins
39200 skills141 agents200 commands
Install
$ npx -y skills add wyre-technology/msp-claude-plugins --agent claude-code

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/portfolio-sweep

Context preview

What this command does when you run it.

Sweep every connected security tool across all clients/tenants, normalize findings, and report the top most urgent items portfolio-wide

Command definition

portfolio-sweep.md
description: Sweep every connected security tool across all clients/tenants, normalize findings, and report the top most urgent items portfolio-wide
argument-hint: ""
arguments: []

Portfolio Security Sweep

Sweep all connected EDR/MDR/SIEM/identity security tools across every client in the portfolio, normalize severity across vendors, and report the most urgent items first — a portfolio-wide equivalent of opening every client's security console at once and reading only what matters.

Prerequisites

  • Conduit gateway connected (`conduit`) with at least one connected security

vendor for at least one client

  • No specific vendor is required — the sweep discovers and adapts to

whatever is connected per client

Steps

1. **Discover available security tools.** Call `conduit__search_tools` to determine which security vendors (EDR, MDR, SIEM, CIPP/identity, email security, SaaS security) are connected, and for which clients. Do not assume any specific vendor from the pack's description — the actual connected set varies per organization and per client.

2. **Enumerate the client/tenant list.** Pull the client list from the PSA or the gateway's tenant listing so every client with any connected security tooling is included in scope.

3. **Query each connected vendor for each client.** For every client, query its connected vendors' current open findings/alerts/incidents/threats. Record which vendors were queried per client and which vendor families had no connector present for that client.

4. **Normalize severity.** Apply the alert-severity-normalization skill's Critical/High/Medium/Low model to every finding pulled, regardless of source vendor, so results are comparable across the portfolio.

5. **Rank and report.** Sort the combined, normalized finding set portfolio-wide by severity (Critical first), then group the top N by client so the reader can see both the overall urgency ranking and which clients are contributing the most risk.

6. **Report coverage gaps.** For every client with no connected security tooling at all, or with a partial vendor mix, state that explicitly — a client with nothing to report because nothing is connected must never look identical to a client with a genuinely clean sweep.

Parameters

This command takes no arguments — it always sweeps the full portfolio and every vendor `conduit__search_tools` reports as connected.

Examples

/secops-pack:portfolio-sweep

Output

# Portfolio Security Sweep
**Run date:** [date]  |  **Clients swept:** [N]  |  **Clients with no connected security tooling:** [N]

## Top Urgent Items (Portfolio-Wide)

| Rank | Severity | Client | Vendor | Finding |
|------|----------|--------|--------|---------|
| 1 | Critical | [client] | [vendor] | [finding summary] |

## By Client

### [Client Name]
Vendors queried: [list]  |  Not connected: [list, or "none"]

| Severity | Vendor | Finding |
|----------|--------|---------|
| Critical | [vendor] | [finding] |

*Repeat per client, ordered by highest severity present.*

## Coverage Gaps

| Client | Missing Vendor Families | Impact |
|--------|--------------------------|--------|
| [client] | [e.g. "No EDR, no SIEM — CIPP alert queue only"] | [what could not be checked] |

Related Commands

  • `/secops-pack:tenant-exposure` — a risk-ranked view (open findings + MFA

gaps + coverage gaps) rather than a raw finding sweep

  • `/secops-pack:incident-report` — build a full timeline once a specific

finding from this sweep needs to become an incident report

Read more
Ships withmsp-claude-plugins

One command to supercharge Claude Code for MSP workflows. Then restart Claude Code. That's it. Documentation: mcp.wyre.ai

Get the whole plugin, auto-invoked
Stats
39
Stars
0
Views
17
Forks
Active
Maintenance
Astro
Language
Apache-2.0
License
1d ago
Last commit
6mo ago
Created

Repo: wyre-technology/msp-claude-plugins