defender-for-storage

Name: defender-for-storage
Author: vinayaklatthe

161stars34forks

How to install defender-for-storage?

vinayaklatthe/microsoft-security-skills/defender-for-storage

$npx -y skills add vinayaklatthe/microsoft-security-skills --skill defender-for-storage

Installs into the current project.

Summary

Guidance for Microsoft Defender for Storage — threat protection for Azure Storage accounts (Blob, Files, Data Lake Gen2). Covers the v2 (per-storage-account) plan, on-upload malware scanning powered by Defender Antivirus, sensitive data threat detection (integrated with Purview c

Files · 1

View on GitHub

SKILL.md

1	# Microsoft Defender for Storage
2
3	Defender for Storage protects Azure Blob Storage, Azure Files, and Azure Data Lake Storage
4	Gen2 against malicious uploads, sensitive data exfiltration, and suspicious access patterns.
5	The current generation is Defender for Storage v2, billed per storage account, with
6	optional add-ons for on-upload malware scanning and sensitive data threat detection.
7
8	## When to use
9	Protecting Azure Storage accounts that receive uploads from untrusted or partner sources,
10	host application content (web apps, ML datasets), or contain sensitive data classified by
11	Purview. Use this skill for plan rollout, malware-scan tuning, and incident response on
12	storage alerts.
13
14	Do not use this skill for SQL/Cosmos data plane protection (`defender-for-cloud-hardening`),
15	VM disk-level malware (`defender-for-servers`), or M365 file workloads
16	(`defender-for-office-365`).
17
18	## Plan and add-ons
19
20	\| Component \| What it does \| Cost model \|
21	\|---\|---\|---\|
22	\| Defender for Storage v2 (base) \| Activity monitoring: anomalous access, suspicious extraction, unusual deletion \| Per storage account / month \|
23	\| On-upload malware scanning \| Defender AV scans every uploaded blob up to 2 GB; result tag + Event Grid event \| Per GB scanned (capped per-account configurable) \|
24	\| Sensitive data threat detection \| Integrates with Purview sensitive-info-types; alerts when sensitive blobs are accessed by suspicious principals \| Included in v2 (no extra fee) \|
25
26	> v1 (per-transaction) is deprecated. Migrate any v1-enabled accounts to v2 — it's
27	> cheaper and predictable.
28
29	## Approach
30
31	1. Enable Defender for Storage v2 at subscription scope with malware scanning on
32	and a per-account monthly GB cap (default 5,000 GB). Cap prevents runaway costs on
33	high-throughput accounts.
34	*Verify: Defender for Cloud → Environment settings → Subscription → Storage shows v2 plan
35	On with malware scanning enabled.*
36
37	2. Override per storage account for outliers:
38	- Disable malware scan on append-only logging accounts, large telemetry sinks, or
39	accounts already scanning at the application layer.
40	- Disable plan entirely on tenant-scoped diagnostic accounts (Activity Logs export,
41	NSG flow logs) where threat detection adds no value.
42
43	3. Wire scan results to Event Grid. Set up an Event Grid system topic per protected
44	account; subscribe a Function/Logic App that:
45	- Moves malicious blobs to a quarantine container (or deletes).
46	- Writes the verdict to the blob's index tag (`Malware-Scanning-scan-result-tag`).
47	- Notifies the SOC.
48	*Verify: upload an EICAR test file; within 2 minutes the blob is tagged Malicious and the
49	quarantine workflow fires.*
50
51	4. Sensitive data threat detection — turn on after Purview classifies the accounts.
52	Without classification, this just runs on file extensions/header heuristics. Pair with
53	Purview Data Map for full value (see `purview-data-map`).
54
55	5. Tune alerts. Baseline 7 days then suppress noisy ones (e.g., expected SAS-token use
56	from CDN edge IPs). Send remaining alerts into Sentinel via the Defender for Cloud
57	connector.
58
59	6. Block public anonymous access at the storage account. Defender alerts on suspicious
60	anonymous reads, but the right control is `allowBlobPublicAccess = false` and Azure Policy
61	to enforce. Defender is detection, not prevention.
62
63	7. Response runbook for "Malicious blob uploaded" alert:
64	1. Confirm the verdict tag on the blob.
65	2. Move to quarantine container; revoke any SAS/keys for the upload identity.
66	3. Hunt sibling uploads from same principal in the past 24h via KQL on
67	`StorageBlobLogs`.
68	4. Notify owner; close the alert with evidence.
69
70	## Guardrails
71	- Always set a per-account GB cap on malware scanning. Without it, a misbehaving client
72	uploading the same 50 GB file 1,000 times can produce a si

People who liked this also viewed

6 picks

Type

defender-for-storage

161stars34forks

View on GitHub

How to install defender-for-storage?

vinayaklatthe/microsoft-security-skills/defender-for-storage

$npx -y skills add vinayaklatthe/microsoft-security-skills --skill defender-for-storage

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Microsoft Defender for Storage
2
3	Defender for Storage protects Azure Blob Storage, Azure Files, and Azure Data Lake Storage
4	Gen2 against malicious uploads, sensitive data exfiltration, and suspicious access patterns.
5	The current generation is Defender for Storage v2, billed per storage account, with
6	optional add-ons for on-upload malware scanning and sensitive data threat detection.
7
8	## When to use
9	Protecting Azure Storage accounts that receive uploads from untrusted or partner sources,
10	host application content (web apps, ML datasets), or contain sensitive data classified by
11	Purview. Use this skill for plan rollout, malware-scan tuning, and incident response on
12	storage alerts.
13
14	Do not use this skill for SQL/Cosmos data plane protection (`defender-for-cloud-hardening`),
15	VM disk-level malware (`defender-for-servers`), or M365 file workloads
16	(`defender-for-office-365`).
17
18	## Plan and add-ons
19
20	\| Component \| What it does \| Cost model \|
21	\|---\|---\|---\|
22	\| Defender for Storage v2 (base) \| Activity monitoring: anomalous access, suspicious extraction, unusual deletion \| Per storage account / month \|
23	\| On-upload malware scanning \| Defender AV scans every uploaded blob up to 2 GB; result tag + Event Grid event \| Per GB scanned (capped per-account configurable) \|
24	\| Sensitive data threat detection \| Integrates with Purview sensitive-info-types; alerts when sensitive blobs are accessed by suspicious principals \| Included in v2 (no extra fee) \|
25
26	> v1 (per-transaction) is deprecated. Migrate any v1-enabled accounts to v2 — it's
27	> cheaper and predictable.
28
29	## Approach
30
31	1. Enable Defender for Storage v2 at subscription scope with malware scanning on
32	and a per-account monthly GB cap (default 5,000 GB). Cap prevents runaway costs on
33	high-throughput accounts.
34	*Verify: Defender for Cloud → Environment settings → Subscription → Storage shows v2 plan
35	On with malware scanning enabled.*
36
37	2. Override per storage account for outliers:
38	- Disable malware scan on append-only logging accounts, large telemetry sinks, or
39	accounts already scanning at the application layer.
40	- Disable plan entirely on tenant-scoped diagnostic accounts (Activity Logs export,
41	NSG flow logs) where threat detection adds no value.
42
43	3. Wire scan results to Event Grid. Set up an Event Grid system topic per protected
44	account; subscribe a Function/Logic App that:
45	- Moves malicious blobs to a quarantine container (or deletes).
46	- Writes the verdict to the blob's index tag (`Malware-Scanning-scan-result-tag`).
47	- Notifies the SOC.
48	*Verify: upload an EICAR test file; within 2 minutes the blob is tagged Malicious and the
49	quarantine workflow fires.*
50
51	4. Sensitive data threat detection — turn on after Purview classifies the accounts.
52	Without classification, this just runs on file extensions/header heuristics. Pair with
53	Purview Data Map for full value (see `purview-data-map`).
54
55	5. Tune alerts. Baseline 7 days then suppress noisy ones (e.g., expected SAS-token use
56	from CDN edge IPs). Send remaining alerts into Sentinel via the Defender for Cloud
57	connector.
58
59	6. Block public anonymous access at the storage account. Defender alerts on suspicious
60	anonymous reads, but the right control is `allowBlobPublicAccess = false` and Azure Policy
61	to enforce. Defender is detection, not prevention.
62
63	7. Response runbook for "Malicious blob uploaded" alert:
64	1. Confirm the verdict tag on the blob.
65	2. Move to quarantine container; revoke any SAS/keys for the upload identity.
66	3. Hunt sibling uploads from same principal in the past 24h via KQL on
67	`StorageBlobLogs`.
68	4. Notify owner; close the alert with evidence.
69
70	## Guardrails
71	- Always set a per-account GB cap on malware scanning. Without it, a misbehaving client
72	uploading the same 50 GB file 1,000 times can produce a si

People who liked this also viewed

6 picks

Type