# Cloud configuration: which settings let attackers in, and would you notice if one changed?

This month, run CISA's ScubaGear against Microsoft 365 and a free posture check against each cloud, and fix public storage, long-lived access keys, open app consent and open external access first. Then lock those settings with organization-level guardrails and re-run the same checks every week, so a changed setting shows up as a diff rather than as an incident.

Version 1.1. Reviewed 5 Nov 2026, next review 5 Dec 2026.
Latest version: https://sheets.einblick.dev/cloud-configuration/

## If you only do five things

1. **Measure against one published baseline this month.** Run CISA's free ScubaGear against Microsoft 365; it only reads settings[1]. Defender for Cloud's free Foundational CSPM checks Azure, AWS and Google Cloud against Microsoft's benchmark[12].
2. **Block public storage at the top.** Enforce S3 Block Public Access for the whole AWS organization[16], set `AllowBlobPublicAccess` to false on every Azure storage account[13] and enforce `storage.publicAccessPrevention` in Google Cloud[22]. Data stores left online without access controls remain among the most common errors in the DBIR[6].
3. **Replace long-lived keys with short-lived credentials.** AWS traces a large portion of security events to compromised long-term access keys[17], and in late 2025 attackers broke in with AWS keys suspected to have leaked in 2022[5]. Move people to federated sign-in and workloads to roles or managed identities, then delete the keys.
4. **Close app consent and external access in Microsoft 365.** Restrict user consent to verified publishers and low-risk permissions, with an admin consent workflow[2]. Allow Teams external access only for named domains: confirmed vishing over cross-tenant Teams calls rose 502% in a year[7].
5. **Turn the baseline into guardrails, then diff it weekly.** Enforce what must never change with AWS Organizations policies, Azure Policy and Google Cloud organization policies, then re-run the assessment weekly and compare runs[1].

## Adopt a baseline without breaking things

1. **Pick one baseline per platform.** For Microsoft 365, SCuBA, or Baseline Security Mode, which every plan has[8]; for Azure, the Microsoft cloud security benchmark[12]; for AWS, Foundational Security Best Practices[15]; for Google Cloud, its security baseline constraints[21].
2. **Assess, and write down each exception.** Export every failing check. Record why each one stays, who owns it and when it ends; ScubaGear's YAML config holds these as exclusions, annotations or omissions[1].
3. **Check the impact, then fix.** Start with the policies CISA marks as BOD 25-01 requirements[4], and turn on each Baseline Security Mode setting whose impact report shows zero impact[8]. For MFA settings, use the passkeys and MFA sheet.
4. **Enforce at the top of each cloud.** In AWS, SCPs cap what roles can do, RCPs what resources allow, and declarative policies set defaults such as IMDSv2[19][18]. In Azure, assign deny policies at a management group. In Google Cloud, add the baseline constraints if your organization predates 3 May 2024[21]. Test each on one account first.
5. **Diff every week.** Compare each run with the last, for example with ScubaGear's `Invoke-SCuBADiff`[1]. Scan infrastructure-as-code pull requests with Checkov or a similar checker[23], and treat a scheduled `terraform plan` that isn't empty as drift.

## Settings attackers go after

- **Public storage.** Google's public access prevention revokes existing public access once you enforce it[22]. Move public sites and downloads to a separate account or project first.
- **Access keys and shared keys.** Attackers reach Azure Blob Storage with exposed credentials, shared keys and SAS tokens[7]; accounts accept keys until `AllowSharedKeyAccess` is false[14]. Google Cloud blocks service account keys in organizations created since May 2024[21] and disables exposed ones[5].
- **Instance metadata.** Require IMDSv2, which uses token-backed sessions, by account default or declarative policy. Track the `ImdsV1Enabled` and `MetadataNoToken` metrics to zero, then add `ec2:RoleDelivery` to an SCP so IMDSv1 credentials stop working[18].
- **Broad roles for workloads.** Remove wildcard actions (Security Hub controls IAM.1 and IAM.21)[15] and pin GitHub OIDC trust to one repository and branch or environment[20]. An over-broad deploy role took UNC6426 from a stolen GitHub token to AWS administrator in under 72 hours[5].
- **Guests and external access.** Let only admins and Guest Inviters invite guests, and only from approved domains[2]; the external sharing sheet explains how this now also limits SharePoint sharing. Allow Teams external domains one by one and block chats from unmanaged accounts. Google Cloud's `iam.allowedPolicyMemberDomains` does the same for IAM grants[21]. SharePoint links: see the DLP sheet.
- **OAuth app consent.** Microsoft's managed default for new tenants still lets users consent to permissions outside a list of mail, file, calendar and chat scopes[10]. Changes keep existing grants[9], so review those too. Stolen Salesloft Drift tokens let attackers bulk-export Salesforce data[5].

## What changed in 2026

- **Exploits overtook weak credentials in Google Cloud.** In late 2025, software flaws opened 44.5% of Google Cloud intrusions in Google's data and weak or missing credentials 27.2%; misconfiguration fell to 21%. Google suggests its secure defaults may have driven the shift[5]. Patch within the patching sheet's deadlines.
- **ScubaGear 2.0 is out.** The October 2026 release adds a Security Suite baseline for Defender, Purview, Exchange Online and Teams[3], so replace `defender` with `securitysuite` in your config[1]. The Entra baseline now requires blocking risky AI agents where Agent 365 or E7 is licensed[2].
- **Microsoft 365 moved toward secure defaults.** Baseline Security Mode, in the admin center since November 2025, covers phishing-resistant MFA for admins, legacy authentication and app consent[8]. Tenant Configuration Management, generally available since April 2026, monitors over 200 settings in six workloads[11].
- **S3 public access can be blocked for a whole organization.** Since November 2025 one AWS Organizations policy blocks it in every account, including new ones[16].

## What you can skip

- **A paid CSPM before the free checks.** ScubaGear and Defender for Cloud's Foundational CSPM cost nothing[1][12]. Buy attack path analysis once the basics are closed.
- **Perfecting 90-day key rotation.** Security Hub checks for it[15], but AWS advises not creating long-term keys at all[17]. Remove them instead.
- **A home-grown baseline.** Start from a published one that someone keeps current.

## Mistakes that cost the most

- **Trusting a new default to fix old resources.** Enforcing IMDSv2 for an account leaves instances that already allow IMDSv1 alone[18], and consent changes keep old grants[9].
- **Running workloads in the AWS management account.** SCPs and RCPs don't restrict it[19]. Keep it for billing and organization tasks.
- **Exposing a test resource "just for a day."** Microsoft measured a median of 5.3 hours from an exposed container's start to the first exploit attempt[7].

## What this looks like at your size

### Small team, no security staff

- Turn on each Baseline Security Mode setting with zero impact[8].
- Ask your IT provider for a monthly ScubaGear report and the list of consented apps and guests.
- Block public storage for the whole cloud organization, not account by account.

### Growing, with IT or security staff

- Run ScubaGear weekly from a scheduled job, guard its app certificate like an admin credential and post the diff to the team channel.
- Connect every cloud account to one posture tool and give each finding an owner and a due date.
- Make infrastructure as code the only write path, and keep console write access for emergency roles.

### Enterprise or regulated

- Treat BOD 25-01's mandatory SCuBA policies as your floor, as US federal civilian agencies must[4].
- Feed drift and posture findings into the SIEM, with a fix time per severity.
- Review changes to guardrail policies like Tier 0 code, with two people.


## Version history

- 5 Nov 2026, version 1.1: Noted that, since Entra B2B integration became permanent for SharePoint in July 2026, limiting guest invitations also limits SharePoint and OneDrive sharing with new outsiders; the external sharing sheet has the details.
- 4 Nov 2026, version 1.0: Published after a copy edit and a source-by-source fact check, which corrected the Salesloft Drift example and the IMDSv1 cutover order and narrowed 4 statements to what the sources say.
- 4 Nov 2026, version 0.1: First draft, based on CISA's SCuBA baselines and BOD 25-01, Google's Cloud Threat Horizons Report H1 2026, the 2026 DBIR, the Microsoft Digital Defense Report 2026 and AWS, Microsoft Learn and Google Cloud documentation.

## Sources

1. ScubaGear (README and docs). CISA, v2.0.0, Oct 2026. https://github.com/cisagov/ScubaGear
2. CISA M365 Secure Configuration Baseline for Microsoft Entra ID. CISA (ScubaGear), ScubaGear v2.0.0, Oct 2026. https://github.com/cisagov/ScubaGear/blob/main/PowerShell/ScubaGear/baselines/aad.md
3. CISA Publishes Significant Updates to ScubaGear and ScubaGoggles. CISA, 27 Oct 2026. https://www.cisa.gov/news-events/news/cisa-publishes-significant-updates-scubagear-and-scubagoggles
4. BOD 25-01: Implementing Secure Practices for Cloud Services. CISA, 17 Dec 2024. https://www.cisa.gov/news-events/directives/bod-25-01-implementing-secure-practices-cloud-services
5. Cloud Threat Horizons Report H1 2026. Google Cloud, Mar 2026. https://services.google.com/fh/files/misc/cloud_threat_horizons_report_h12026.pdf
6. 2026 Data Breach Investigations Report. Verizon, 19 May 2026. https://www.verizon.com/business/resources/T6d3/reports/2026-dbir-data-breach-investigations-report.pdf
7. Microsoft Digital Defense Report 2026. Microsoft Threat Intelligence, Oct 2026. https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Microsoft-Digital-Defense-Report-2026-v1.1.pdf
8. Baseline security mode settings. Microsoft Learn, 30 Sep 2026. https://learn.microsoft.com/en-us/microsoft-365/baseline-security-mode/baseline-security-mode-settings
9. Configure how users consent to applications. Microsoft Learn, 15 Jun 2025. https://learn.microsoft.com/en-us/entra/identity/enterprise-apps/configure-user-consent
10. Manage app consent policies. Microsoft Learn, 22 Jan 2026. https://learn.microsoft.com/en-us/entra/identity/enterprise-apps/manage-app-consent-policies
11. Tenant Configuration Management APIs are now generally available. Microsoft Entra Blog, 23 Apr 2026. https://techcommunity.microsoft.com/blog/microsoft-entra-blog/tenant-configuration-management-apis-are-now-generally-available/4513157
12. What is Cloud Security Posture Management (CSPM). Microsoft Learn, 11 Sep 2026. https://learn.microsoft.com/en-us/azure/defender-for-cloud/concept-cloud-security-posture-management
13. Remediate anonymous read access to blob data. Microsoft Learn, 15 Jul 2026. https://learn.microsoft.com/en-us/azure/storage/blobs/anonymous-read-access-prevent
14. Prevent Shared Key authorization for an Azure Storage account. Microsoft Learn, 11 Aug 2026. https://learn.microsoft.com/en-us/azure/storage/common/shared-key-authorization-prevent
15. AWS Foundational Security Best Practices standard in Security Hub CSPM. AWS Documentation, checked 4 Nov 2026. https://docs.aws.amazon.com/securityhub/latest/userguide/fsbp-standard.html
16. Amazon S3 Block Public Access now supports organization-level enforcement. AWS, 26 Nov 2025. https://aws.amazon.com/about-aws/whats-new/2025/11/amazon-s3-block-public-access-organization-level-enforcement/
17. Preventing unintended encryption of Amazon S3 objects. AWS Security Blog, updated 18 Mar 2025. https://aws.amazon.com/blogs/security/preventing-unintended-encryption-of-amazon-s3-objects/
18. Transition to using Instance Metadata Service Version 2. AWS Documentation, checked 4 Nov 2026. https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instance-metadata-transition-to-version-2.html
19. Managing organization policies with AWS Organizations. AWS Documentation, checked 4 Nov 2026. https://docs.aws.amazon.com/organizations/latest/userguide/orgs_manage_policies.html
20. Configuring OpenID Connect in Amazon Web Services. GitHub Docs, checked 4 Nov 2026. https://docs.github.com/en/actions/how-tos/secure-your-work/security-harden-deployments/oidc-in-aws
21. Manage Google Cloud security baseline constraints. Google Cloud, updated 26 Oct 2026. https://docs.cloud.google.com/organization-policy/manage-baseline-constraints
22. Organization policy constraints. Google Cloud, updated 26 Oct 2026. https://docs.cloud.google.com/organization-policy/reference/org-policy-constraints
23. What is Checkov?. Checkov docs, checked 4 Nov 2026. https://www.checkov.io/1.Welcome/What%20is%20Checkov.html
