Exchange Online Protection Checker
Use it alongside Configuration analyzer in the Defender portal, which compares the same values live.
Free
Works with the AI Agents you already use
Exchange Online Protection Checker
Example session with this skill installed
Is our email protection configured right? We have Exchange Online Plan 1 only (no Defender for Office 365). Here is Get-HostedContentFilterPolicy -Identity Default | fl, Get-HostedOutboundSpamFilterPolicy | fl and Get-MalwareFilterPolicy | fl. SPF for example.com: v=spf1 include:spf.protection.outlook.com ~all. No DMARC record yet.
- Read your context and instructions
- Compiled the exchange online protection
Illustrative example with made-up values for a fictional domain (example.com).
Email protection check, 28 October 2026
Licence: EOP · Policies read: 3 · Settings compared: 16 (match 10 · weaker 5 · not set 1)
| Severity | Policy | Setting | Current | Target | Fix |
|---|---|---|---|---|---|
| Critical | Outbound (Default) | AutoForwardingMode | Automatic | Off | Defender portal > Threat policies > Anti-spam > outbound policy > Forwarding rules |
| Critical | Anti-spam (Default) | HighConfidencePhishAction | MoveToJmf | Quarantine | Set-HostedContentFilterPolicy (HighConfidencePhishAction) |
| Critical | example.com | DMARC | none | p=none with rua, then p=reject | Publish _dmarc TXT record |
| High | Anti-spam (Default) | PhishZapEnabled | False | True | Set-HostedContentFilterPolicy (PhishZapEnabled) |
| High | Outbound (Default) | ActionWhenThresholdReached | Alert | BlockUser | Outbound policy > Restriction placed on users |
| Medium | Anti-spam (Default) | BulkThreshold | 7 | 6 | Set-HostedContentFilterPolicy (BulkThreshold) |
User impact: external auto-forwards get NDR 5.7.520; high confidence phishing goes to admin-only quarantine instead of Junk.
Rollback: set the previous values from the pasted output.
DMARC and authentication
| Domain | SPF | DKIM | DMARC today | Next step |
|---|---|---|---|---|
| example.com | ~all (Microsoft recommends -all) | not assessed | missing | p=none; pct=100; rua=mailto:reports mailbox, read reports 2 to 4 weeks |
Not assessed
Anti-phishing spoof and DMARC handling (paste Get-AntiPhishPolicy), DKIM (paste Get-DkimSigningConfig). Safe Links, Safe Attachments and impersonation need Defender for Office 365 Plan 1.
Summary for management
Ten of sixteen settings match Microsoft's recommended values. The biggest gap is that users can still auto-forward mail outside the company and high confidence phishing lands in Junk. This week we block external forwarding, quarantine phishing and start DMARC monitoring; users only notice an NDR if they forward mail outside.
Connects securely to your tools. The creator never sees your data.
About this skill
Paste the output of the Exchange Online protection cmdlets (anti-spam, outbound spam, anti-phishing, anti-malware, Safe Links, Safe Attachments, DKIM) or your SPF and DMARC records, and the skill:
- compares every setting it can read with Microsoft's Standard (or Strict) recommended values
- marks each one match, weaker, stronger or not set, and ranks the gaps: critical, high, medium
- gives a fix list of up to 10 items with the target value, where to change it, the user impact and the rollback
- builds a DMARC plan per domain from p=none to p=reject, subdomains first
- says which checks need Defender for Office 365 Plan 1 or Plan 2
- ends with a 3-sentence summary for management
It never runs commands that change the tenant. Markdown only: no scripts, no network access. Use it alongside Configuration analyzer in the Defender portal, which compares the same values live.
Reference values come from Microsoft Learn, "Recommended settings for EOP and Microsoft Defender for Office 365 security", re-read on 28 October 2026. If Microsoft changes a value, tell the skill and give the Learn page; it uses your value and says so in the report.
Not affiliated with or endorsed by Microsoft.
Requires manual input of PowerShell output or CSVs. Cannot directly execute tenant changes. Defender-specific settings are not assessed unless the license level is provided.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 20 days ago
- Free with an Agensi account