SuperAPI AI Directory

defender-tvm

Security & auditing

View repository

Updated 10-16 09:59

About this skill

Microsoft Defender TI & Vulnerability Management

This skill covers two complementary capabilities:

  • Microsoft Defender Vulnerability Management (MDVM) — risk-based vulnerability management for endpoints (Windows, macOS, Linux, mobile), firmware, browser extensions, certificates, and security baselines.
  • Microsoft Defender Threat Intelligence (Defender TI / MDTI) — Microsoft's finished-intel and raw-IOC product for adversary tracking, infrastructure pivoting, and threat hunting enrichment.

Both feed Defender XDR and Sentinel.

When to use

  • Prioritizing patching/remediation across a large endpoint estate.
  • Measuring exposure score and security baseline drift.
  • Hunting and responding to threats with adversary intel and infrastructure data.
  • Enriching SIEM alerts with intel profiles and indicator context.

Do not use this skill for endpoint policy/onboarding (defender-for-endpoint), SIEM detection rules (sentinel-detection-engineering), or outside-in attack surface (defender-easm).

MDVM — what's in core MDE Plan 2 vs the add-on

Capability MDE Plan 2 MDVM Add-on
CVE-based vulnerability inventory ✅ ✅
Exposure score, recommendations ✅ ✅
Threat-aware prioritization ✅ ✅
Security baselines assessment (CIS, STIG) — ✅
Browser extension inventory — ✅
Digital certificate inventory — ✅
Network share configuration assessment — ✅
Hardware & firmware inventory — ✅
Block vulnerable applications — ✅
Authenticated scan for unmanaged Windows — ✅

Rule of thumb: If you're already on MDE P2 and your audit team asks for CIS-aligned evidence, you need the MDVM add-on.

Approach

MDVM workflow

  1. Triage by exposure score, not CVE count. Exposure score weights exploitability + asset criticality + active threat campaigns. A 5,000-CVE backlog sorted by exposure becomes a 50-task action list.

  2. Use threat insights filters. Microsoft tags CVEs with: Public Exploit, Verified by Microsoft, Active Threat Campaign, Exploit Available in Kit. Patch Active Threat Campaign first, regardless of CVSS.

  3. Wire remediation to Intune. "Create remediation request" pushes a task to Intune admin queue with the affected device list — the patch lifecycle stays in one place.

  4. Security baselines (Plan 2 add-on). Pick CIS or STIG, scope to a device group, review compliance per setting, prioritize the security-impacting settings.

  5. Block vulnerable apps (add-on). Use sparingly: blocks Adobe Reader 21.x, Chrome <120, etc. across the estate. Pilot on a small group; some users have legitimate version pins.

  6. Cadence. Weekly review of top-10 exposure-reducing actions; monthly executive exposure-score trend; quarterly baseline drift report.

Defender TI workflow

  1. Intel profiles — track named adversaries (e.g., Storm-####), threat tools, CVEs, campaigns. Subscribe relevant ones; updates surface in Defender XDR.

  2. Infrastructure pivoting. Given an indicator (IP, domain, hash), Defender TI shows pDNS, WHOIS history, certificate associations, related infrastructure — pivot to find adversary infrastructure your SIEM hasn't seen yet.

  3. Bring intel into hunting.

    • Defender XDR advanced hunting: indicators are first-class joinable entities.
    • Sentinel: connect MDTI via the Threat Intelligence Premium data connector; ThreatIntelligenceIndicator and articles populate.
  4. MDTI API for automation. Pull articles/indicators into a SOAR playbook for auto-enrichment of incidents.

  5. Free vs Premium. Defender TI portal has a free tier (limited articles, pivots/day); Premium unlocks unlimited pivots, full intel-profile catalogue, and API throughput. Most enterprise SOC use needs Premium.

Guardrails

  • Don't drown in CVE counts. Exposure score + threat insight is the actionable view. Counting CVEs is a vanity metric.
  • MDVM remediation needs an owner. Surfacing 200 recommendations in a dashboard with no Intune integration / patch owner produces no patches.
  • Block-vulnerable-apps is invasive. Pilot. Communicate. Some apps must be pinned for business reasons.
  • Security baselines are a journey. 100% CIS compliance is rarely achievable or desirable; pick the security-impacting subset.
  • Intel without context is noise. Don't ingest every MDTI indicator into a Sentinel watchlist as a "match-and-alert" — match-rate × low fidelity = alert fatigue.
  • Confirm licensing. MDTI Premium and the MDVM add-on are separately licensed; customers often think they're included in M365 E5 — they aren't.
  • Scan results lag by hours; don't expect minute-by-minute vuln status.

Common anti-patterns

  • "Patched 1,000 medium CVEs while a critical exploited CVE sat for 60 days" — sort by threat campaign, not CVSS bucket.
  • "Bought MDVM add-on for the exposure score we already had in MDE P2" — overlap. Map features before buying.
  • "All MDTI indicators auto-blocked at the firewall" — false-positive blast. Pivot/validate first.
  • "Security baseline 'Compliance' score treated as security score" — they're not the same; many baseline settings are operational hardening, some are not security-critical.
  • "Intel profile subscriptions = no triage" — articles need analyst triage to translate to detections/blocks.
  • "Authenticated scan on every unmanaged Windows host without scoping" — generates service-account auth events at scale; coordinate with identity team.

Example prompts

  • Top 10 actions to reduce our exposure score by 20% this quarter.
  • Patch prioritization runbook: how to use threat insights to triage 8,000 open CVEs.
  • Roll out CIS Windows baseline assessment across 30,000 endpoints.
  • Wire MDVM remediation requests to Intune and an ITSM ticket queue.
  • Pivot from a suspicious domain to related adversary infrastructure using MDTI.
  • Connect MDTI Premium to Sentinel and build a hunting workbook.
  • Pilot block-vulnerable-apps for outdated Java runtimes on dev machines.
  • Build a quarterly executive exposure-score and intel-profile briefing.

Microsoft Learn