Securitain Security Intelligence Hub
Understand How Cloud & AI Attacks Happen — Before They Become Your Incident
Explore real attack techniques, AWS security risks, AI-agent threats, and practical defensive controls. Build executive-ready security briefings using free interactive tools.
MITRE ATT&CK Enterprise v20.1MITRE ATLAS v2026.10No sign-up · runs in your browser
Executive threat overview
The cloud and AI risks worth a leadership conversation
Cloud identity threats
How valid credentials, federation and trust become the main way into AWS.Credential compromise
What a single leaked access key can reach — and how to limit it.Privilege escalation
Permission combinations that quietly turn developers into administrators.Cross-account exposure
When another account’s compromise becomes yours.Security logging impairment
How attackers switch off CloudTrail — and the guardrails that stop them.Data collection & exfiltration
S3 reads, snapshot sharing and data leaving through AWS itself.AI-agent risks
Agents with broad permissions, prompt injection and unapproved actions.Kubernetes & workload identity
From a compromised pod to cluster secrets and cloud data.
Free interactive utilities
Tools you can use today — no account, nothing uploaded
Cloud Attack Scenario Explorer
Twenty interactive AWS and AI attack paths with controls, telemetry and business impact.Free · runs in your browserExecutive Cyber Risk Brief Builder
Answer 18 questions and get a board-ready, self-reported readiness brief — generated in your browser.Free · runs in your browserAI Agent Security Self-Assessment
Assess an AI agent’s permissions, untrusted-content handling, tool safety and auditability.Free · runs in your browserCloud Control Explorer
Map cloud attack techniques to AWS controls, with the evidence needed to validate them.Free · runs in your browserCloud Logging Readiness Checklist
Check AWS CloudTrail coverage, integrity and monitoring — with practical recommendations.Free · runs in your browser
Browse by security domain
Start from the area you are responsible for
AWS Security
Identity & Access Management
Cloud Data Security
Kubernetes & Containers
AI Agents & LLM Security
CI/CD & Supply Chain
Detection & Logging
Executive Risk & Governance
Framework explorer
Two frameworks, kept deliberately separate
MITRE ATT&CK Enterprise
Adversary behavior across enterprise, cloud and identity environments.
- Version
- 20.1
- Tactics
- 15
- Techniques
- 227 + 479 sub
Imported from the official STIX release (source modified 2026-11-05).
MITRE ATLAS
Adversary behavior against AI-enabled systems, with real-world case studies.
- Version
- 2026.10
- Tactics
- 16
- Techniques
- 129
Imported from the official release (source modified 2026-10-28). ATLAS IDs (AML.*) are never treated as ATT&CK IDs.
Featured attack scenarios
Interactive attack paths with business impact
- Data exposurePersistent access
Compromised AWS access key
A long-lived IAM user access key leaks from a developer laptop or repository and is used to explore the account, add persistence and read data.Business impact: Customer or internal data in reachable buckets may be copied; without data events the scope may be impossible to determine precisely.5 stages - Data exposurePrivilege takeover
Dangerous cross-account role trust
A role trusts an entire external account without conditions. When that account is compromised, the adversary walks straight into yours.Business impact: Data reachable by the role can be read by whoever controls the trusted account.4 stages - Privilege takeoverData exposure
Privilege escalation through IAM policy relationships
A developer role that can pass roles to Lambda and create functions can run code as a far more privileged role — without ever being granted administrator rights directly.Business impact: Full administrative control of the account allows changes to any resource or identity.5 stages - Loss of visibility
CloudTrail logging impairment
Before acting, an adversary with administrative rights stops or narrows CloudTrail logging so that later activity leaves little evidence.Business impact: Investigation becomes slower and less certain; responders may have to assume worst-case scope.5 stages - Unauthorized actionsPrivilege takeover
AI agent with excessive AWS permissions
An operations agent that can call AWS APIs runs under a broad role. Manipulated input causes it to make changes no human approved.Business impact: Production infrastructure or access can be changed without human intent.4 stages
Executive insights
Questions CTOs, CISOs and founders ask us
- What happens when an AWS access key is compromised?
- How can CloudTrail logging be disabled — and prevented?
- What is indirect prompt injection?
- How should companies audit AI-agent permissions?
- What changed when ATT&CK split Defense Evasion?
- How can a CTO assess AWS security readiness?
More analysis in Securitain Insights.
How to read security coverage
Why we never say “protected against T1098”
Mapping a control to a technique does not prove protection. Wherever coverage is discussed, Securitain distinguishes six states. The educational content on this site describes potential exposures and relevant controls; it does not represent detection capability in your environment.
- Potential exposure
A configuration, identity relationship or permission could create a risky condition.
- Preventive control
A verified safeguard limits feasibility under identified assumptions.
- Detection-ready
Required telemetry and a working detection have been confirmed.
- Observed behavior
Evidence indicates activity consistent with a technique.
- Not assessed
The necessary inspection was not conducted, or evidence is missing.
- Not applicable
The technique is outside the assessed scope.
Methodology
How the hub is built
Framework data
Technique names, IDs, tactics, platforms, mitigations and detection strategies are imported from MITRE’s official, versioned releases and validated before publishing. Revoked and deprecated techniques are labelled, never silently reused.
Securitain analysis
Scenarios, guides and control relationships are written by Securitain, reference only validated technique IDs, and state their assumptions. Scenarios are fictional and labelled as such.
Assessment scoring
Deterministic weighting: yes = 1, partially = 0.5, no = 0; unknown answers lower confidence instead of counting as failures; not-applicable answers are excluded. No AI model scores your answers.
Privacy
Assessments run entirely in your browser. Answers are not stored, not sent to Securitain and not placed in URLs. Optional analytics record only coarse events (for example “assessment completed”) after cookie consent.
Sources and licensing
Attribution
MITRE ATT&CK® data: mitre-attack/attack-stix-data, Enterprise v20.1. © The MITRE Corporation. This work is reproduced and distributed with the permission of The MITRE Corporation. See the ATT&CK terms of use.
MITRE ATLAS™ data: mitre-atlas/atlas-data, v2026.10, licensed under the Apache License 2.0.
MITRE does not endorse Securitain. ATT&CK does not enumerate every possible adversary behavior; using it to cover categories of techniques does not guarantee full defensive coverage.