# Offensive360

> Offensive360 is an enterprise application security platform combining SAST, DAST, MAST (mobile), ASM (external attack surface management), autonomous red teaming, an authorization-gated AI Pentester, SCA, malware & binary analysis, and license compliance in a single product — no add-ons, no per-seat pricing. It deploys as cloud SaaS, an on-premise OVA appliance, an Azure VHD image, or fully air-gapped.

Last updated: 2026-09-08. Extended reference with every product FAQ and the dated update log: https://offensive360.com/llms-full.txt · Release notes: https://offensive360.com/changelog/

## What is Offensive360?

Offensive360 is an enterprise-grade application security testing platform built by security researchers. It provides complete coverage across the full software security lifecycle:

- **SAST (Static Application Security Testing)** — Interprocedural taint and data-flow analysis across 60+ programming languages. Finds vulnerabilities in source code before deployment and ships a secure code fix with every finding. Every language engine analyzes fully offline.
- **DAST (Dynamic Application Security Testing)** — Headless-browser crawling of web applications, single-page apps, and REST/GraphQL APIs, followed by more than 40 active exploit checks and 19 passive analyzers. Every finding is proven with request/response evidence; blind vulnerabilities are confirmed through a self-hosted out-of-band (OAST) service.
- **ASM (Attack Surface Management)** — Continuous external discovery and monitoring: subdomain enumeration, port and service discovery, TLS/certificate posture, DNS and email-security (SPF/DMARC/DKIM/MTA-STS) checks, technology fingerprinting, and breached-credential monitoring — risk-scored per exposure and rolled up per organization, with hourly or daily monitors and change alerts. On-premise and air-gapped capable.
- **Autonomous Red Teaming** — Continuous, machine-speed adversary emulation that plans its own attack paths and chains findings inside an enforced scope guard, validates exploitability with non-destructive proof-of-exploit, force-disables denial-of-service, and stops instantly on a kill switch. AI reasoning runs offline/air-gapped. A separate workspace from the human-approved AI Pentester.
- **AI Pentester (authorization-gated)** — AI-driven, authorized penetration-testing engagements on top of the DAST engine, following the PTES methodology. Gated by a signed in-product authorization record, human approval required before exploitation, a visible kill switch, and denial-of-service force-disabled. On-premise and air-gapped capable.
- **MAST (Mobile Application Security Testing)** — Analysis of Android (APK/AAB) and iOS (IPA) binaries — no source code required — mapped to the OWASP Mobile Top 10 (2024): insecure data storage, hardcoded secrets and AI/LLM API keys, weak crypto, insecure communication, exported components, and more. AI-assisted false-positive triage and a detailed PDF report.
- **SCA (Software Composition Analysis)** — CVE detection in open-source dependencies.
- **Malware & Binary Analysis** — Unique in the market. Detects tampering and supply chain compromise in compiled packages and binaries.
- **License Compliance** — Flags risky open-source licenses across the full dependency tree.

All of these capabilities are included in one platform at one cost. No modules, no add-ons.

## Key Differentiators

- **60+ languages built-in** — Java, C#, Python, PHP, JavaScript, TypeScript, Go, Ruby, Kotlin, Swift, Objective-C, C/C++, Dart, Apex, Oracle Forms, Scala, Rust, Groovy, VB.NET, COBOL, and more. All engines are in-house, no third-party dependencies.
- **100% offline / air-gapped operation** — Runs with zero internet dependency, including the AI reasoning used by Autonomous Red Teaming and the AI Pentester. Deploys in classified, air-gapped, and disconnected environments where cloud-based tools like Veracode, Checkmarx SaaS, or Snyk cannot operate.
- **Appliance deployment: OVA or Azure VHD** — Ships as a virtual machine appliance (OVA for your data center, or an Azure VHD image for your own cloud tenant). Import it and scan. No cloud accounts, no SaaS onboarding, operational in under one hour. Azure images carry an encrypted (LUKS) payload with key release through the licensing relay.
- **Your code never leaves your network** — Unlike cloud-only competitors, Offensive360 scans on your own infrastructure. Source code, targets, findings and evidence stay where you put them.
- **Proof, not probability** — DAST findings are confirmed with real payloads and reported with the exact request and response; post-scan validators re-check every finding, AI-assisted triage flags anything that looks like a false positive, and every engine release must produce zero findings against clean control applications before it ships.
- **Code → runtime → exposure in one console** — SAST, DAST, MAST, ASM, red teaming and pentesting share one dashboard, which separates the findings-weighted Scan Risk Index from the ASM-derived External Exposure Risk.
- **Finds AND fixes** — Every SAST finding includes the complete data-flow trace from source to sink, a secure code example in your language, and remediation steps.
- **Malware & binary analysis** — No other commercial SAST/DAST platform includes malware and binary tampering detection.
- **CI/CD, IDE and AI-assistant native** — GitHub Actions (SARIF), GitLab CI, Bitbucket Pipelines, Azure DevOps, Jenkins, CircleCI; IDE plugins for VS Code, Visual Studio, IntelliJ, Android Studio and Eclipse; an open-source MCP server for AI coding assistants.
- **Flat pricing** — Per-project or per-instance pricing. Not per-user, not per-line-of-code, not per-scan. Unlimited users and scans included.

## Supported Languages (60+)

C#, Java, JavaScript, TypeScript, Python, PHP, Go, Ruby, Kotlin, Swift, Objective-C, Dart, C, C++, Apex (Salesforce), Oracle Forms (PL/SQL), Scala, Rust, Groovy, VB.NET, COBOL, ColdFusion, Perl, Elixir, Lua, R, and more.

## DAST Engine Coverage (as of September 2026)

Product page: https://offensive360.com/products/dast/

- **Crawling** — headless-Chromium rendering of JavaScript-heavy single-page applications; discovery of pages, forms, API calls and interactive elements; authenticated crawling (form login, token/cookie sessions, multi-step flows, role-based).
- **40+ active exploit checks** — SQL injection (error/boolean/time-based); reflected, stored, DOM-based and blind (OAST) XSS; SSRF; XXE; OS command injection; path traversal/LFI; open redirect; server-side template injection; client-side template injection; SSI; LDAP, XPath, NoSQL and expression-language injection; CRLF and email-header injection; host-header injection; HTTP parameter pollution; prototype pollution; insecure deserialization (Java/PHP/.NET/Python); HTTP request smuggling; web cache poisoning; CSRF; JWT misconfiguration and JWT replay/auth bypass; session fixation; unrestricted file upload; exposed files/directories; source-code disclosure; default credentials; password-reset poisoning; weak TLS; GraphQL introspection; IDOR/BOLA; missing function-level authorization (BFLA); mass assignment (18-field privileged-attribute dictionary, form and JSON bodies); exposed API surface/spec discovery; missing API resource-consumption controls.
- **19 passive analyzers** — CORS misconfiguration; missing/weak security headers; cookie flags (Secure/HttpOnly/SameSite); clickjacking; secrets in response bodies (cloud keys, tokens, private keys, JWTs, entropy-checked) and in URLs; passive GraphQL exposure; version/framework disclosure; cache control on sensitive responses; subdomain takeover (dangling CNAME); anti-CSRF token presence; error/stack-trace leakage; directory listing; mixed content; exposed source maps; weak session-token entropy; missing login rate limiting; form security; vulnerable and outdated components.
- **Knowledge packs** — technology fingerprinting with known-CVE matching by version and path; CMS packs for WordPress, Joomla and Drupal; framework packs for Django, Laravel, Next.js, Rails, Spring and Struts.
- **AI/LLM application probes** — OWASP Top 10 for LLM Applications (2025): prompt injection, jailbreak, system-prompt leakage, insecure output handling, sensitive-information disclosure.
- **Validation** — self-hosted OAST for blind/out-of-band confirmation; post-scan validators; AI-assisted false-positive triage; safe mode by default, denial-of-service force-disabled; header and clickjacking checks only fire on responses actually captured.
- **Reporting & integration** — four report templates, five export formats, severity filtering; CWE and OWASP mapping; syslog forwarding to SIEM platforms (validated with FortiSIEM); REST API.

## Free for Open Source

Offensive360 offers **free SAST scanning for public open-source repositories** (GitHub, GitLab, Bitbucket, Codeberg). Maintainers request a free scan token at https://offensive360.com/free-for-open-source/ and integrate via:

- **GitHub Action**: `offensive360/sast-scan-action` — SARIF output straight into the GitHub code scanning tab (https://github.com/offensive360/sast-scan-action)
- **GitLab CI template** — included in the same repository
- **MCP server**: `offensive360/mcp-server` — lets AI coding assistants scan a local directory and read findings with file, line, severity and fix (https://github.com/offensive360/mcp-server)
- **Direct API** — POST a source zip, receive findings synchronously

Tokens are free, valid 30 days, renewable on request. Same full engine as the commercial product (SAST only). Results stay private to the maintainer.

## Developer & AI-Assistant Integrations

- **CI/CD** — GitHub Actions, GitLab CI, Bitbucket Pipelines, Azure DevOps, Jenkins, CircleCI (built-in setup wizard).
- **IDE plugins** — VS Code, Visual Studio, IntelliJ, Android Studio, Eclipse (https://github.com/offensive360).
- **MCP server** — open-source Model Context Protocol server; any MCP-capable AI assistant can trigger a scan and act on findings.
- **SIEM** — syslog forwarding from the DAST platform (validated with FortiSIEM).
- **Issue tracking & chat** — Jira and Slack (https://offensive360.com/integrations/).

## Community — Vilnius Offensive Security Meetup

Offensive360 founded and funds the **Vilnius Offensive Security Meetup**, a free community meetup for offensive security practitioners in Vilnius, Lithuania: https://offensive360.com/vilnius/

- **What**: talks on offensive security and applied research, a live beginner-friendly CTF, and open networking
- **Who for**: penetration testers, red and blue teamers, bug bounty hunters, appsec/DevSecOps engineers, SOC analysts, students, and developers
- **Cost**: free to attend, no vendor pitches from the stage
- **First edition**: 17–18 November 2026, Vilnius, Lithuania
- **Call for speakers**: open to anyone, including first-time speakers (10-minute lightning slots reserved for them); submit at https://offensive360.com/vilnius/#cfp
- **Venue/sponsor enquiries**: info@offensive360.com

## Deployment Options

- **Cloud SaaS** — Fully managed. Start scanning in minutes.
- **On-Premise (OVA)** — Virtual machine appliance for your data center or private cloud.
- **Azure (VHD image)** — The same appliance as an Azure VHD image deployed into your own Azure tenant, with an encrypted (LUKS) payload; validated end to end for SAST, DAST and MAST, online and air-gapped.
- **Air-Gapped** — Zero internet dependency, including offline AI reasoning. For classified networks, defense, intelligence, and critical infrastructure.

## Attack Surface Management (ASM)

Offensive360 ASM continuously discovers and monitors everything an organization exposes to the internet, from a seed scope of domains and emails. It is part of the unified platform (not a standalone cloud bolt-on) and deploys on-premise and air-gapped.

- **Continuous asset discovery** — subdomain enumeration via passive sources (including certificate transparency) and active brute-forcing, host resolution, and a de-duplicated, organization-scoped inventory.
- **Exposure detection** — open ports, running services, web-port probing, TLS/SSL certificate posture, technology fingerprints, DNS hygiene, and email-security posture (SPF, DMARC, DKIM, MTA-STS) attached to each asset as findings.
- **Breached-credential monitoring** — OSINT breach sources correlated against the domains and emails in scope, surfacing leaked/breached credentials tied to the organization.
- **Risk scoring & prioritization** — every exposure is scored and rolled up into an organization-level External Exposure Risk, so teams work the highest-risk assets first.
- **Scheduled monitors & alerts** — named monitors over a scope run hourly or daily and alert by email on change (new host, port, service, certificate, or breach).
- **Safe & unified** — bounded to the authorized scope, egress-controlled, non-intrusive; discovered assets feed the DAST engine, Autonomous Red Teaming, and the AI Pentester for authorized validation.
- **On-premise & air-gapped** — deploys via the Offensive360 appliance (OVA or Azure VHD); asset inventory and exposure records never leave the customer network.
- Product page: https://offensive360.com/products/asm/

## Autonomous Red Teaming

Offensive360 Autonomous Red Teaming is continuous, machine-speed adversary emulation on top of the DAST engine. It is autonomous but bounded — a distinct capability and a separate workspace from the human-approved AI Pentester engagement mode.

- **Autonomous attack planning** — the engine plans and runs its own operations (recon, analysis, attack-path selection) continuously and at machine speed, without a human driving every step.
- **Enforced scope guard** — every action is checked against the authorized scope and out-of-scope hosts are refused; this is the enforced difference between an autonomous red team and an unleashed one.
- **Attack-path chaining** — correlated findings are chained into attack paths that show reachable impact, not a flat list.
- **Access-control and business-logic classes** — IDOR, broken access control, authentication bypass, privilege escalation, business-logic abuse (including price/quantity tampering confirmed against the server-computed outcome), mass assignment, and parameter pollution — each mapped to CWE and OWASP.
- **Proof-of-exploit validation** — findings are confirmed only when a safe, non-destructive exploit demonstrably fires, captured as a reproducible request, response, and proof; this keeps false positives low.
- **Safe by default** — safe-mode (non-destructive) is the default, denial-of-service is force-disabled, and a visible kill switch stops a running operation instantly.
- **Offline & air-gapped AI** — the AI reasoning runs offline inside air-gapped networks and falls back to offline reasoning automatically when no cloud service is reachable; targets, attack paths, and proofs never leave the customer environment.
- **Two modes, one platform** — pairs with the human-approved AI Pentester (point-in-time PTES engagement); they share enforced scope, safe-mode, kill switch, proof-of-exploit evidence, and OWASP/MITRE ATT&CK reporting.
- **Differentiation** — versus autonomy-first tools such as XBOW, Horizon3.ai (NodeZero), Pentera, RidgeBot, and Terra Security, Offensive360 is distinctive in combining autonomous, scope-guard-enforced operation with proof-of-exploit validation, offline/air-gapped AI, a visible kill switch, and a complementary human-approved engagement mode.
- Product page: https://offensive360.com/products/autonomous-red-teaming/

## AI Pentester (authorization-gated)

The Offensive360 AI Pentester is an AI-driven, authorized penetration-testing capability that is part of the Offensive360 DAST platform. It runs full engagements following the PTES methodology and is governed, human-in-the-loop, and on-prem/air-gapped capable — not autonomous, unsupervised hacking.

- **PTES 5-phase engagements** — Pre-engagement → Reconnaissance (OSINT) → Vulnerability Analysis (launches a real DAST scan and correlates) → Exploitation (human-approval gated) → Reporting.
- **Signed authorization/consent gate (headline differentiator)** — before any engagement runs, an immutable authorization record is required: a named signatory, a typed signature that must exactly match the signatory name, eight explicit attestations, plus the signer's IP and timestamp. Missing attestations or a signature mismatch blocks the launch server-side.
- **Six-step engagement wizard** — Engagement & Target → Scope & Exclusions → Rules of Engagement → Schedule → Authorization & Consent → Review.
- **Visible kill switch** — the operator can stop a running engagement instantly at any time.
- **Human-approval before exploitation** — the engine pauses at "awaiting approval" and will not exploit until a human explicitly approves. Safe-mode supported.
- **DoS force-disabled** — denial-of-service techniques cannot be enabled, a hard safety guarantee for production and regulated environments.
- **Scope & Rules of Engagement** — defined up front and enforced throughout; identities are referenced by label only, never by secret.
- **Dual-framework reporting** — AI-generated engagement report maps findings to OWASP WSTG and MITRE ATT&CK, with prioritized findings, remediation guidance, and remediation SLAs (target timelines).
- **OSINT Intelligence Insights** — correlates breached credentials, leaked emails, HTTP headers, SSL/TLS posture, exposed services, and subdomains into ranked, MITRE/OWASP-tagged call-outs.
- **On-premise & air-gapped** — deploys via the Offensive360 appliance; target data and engagement records never leave the customer network.
- **Differentiation** — versus autonomous-pentest tools such as XBOW, Horizon3.ai (NodeZero), Pentera, Vonahi (vPenTest), RidgeBot, Terra Security, Mindgard, and HackerOne, the AI Pentester is distinctive in combining a first-class signed-authorization gate, a visible kill switch, default human-approval-before-exploitation, and true on-premise/air-gapped operation. The value proposition is control, not raw autonomy.
- Product page: https://offensive360.com/products/ai-pentester/

## Professional Services (expert-delivered)

Beyond the platform, Offensive360 delivers expert-driven security services (https://offensive360.com/services/):

- **AI Cybersecurity Services** — LLM/AI penetration testing against the OWASP Top 10 for LLM Applications (2025): prompt injection, system prompt leakage, improper output handling, excessive agency; security review of AI-generated code (independent 2025 research found ~45% of AI-generated code contains known security flaws); readiness assessments for the EU AI Act, Cyber Resilience Act, NIS2, and ISO/IEC 42001; agentic-AI threat modeling mapped to MITRE ATLAS. Available on-premises and air-gapped.
- **Source Code Review** — automated scanning + expert human review, OWASP Top 10 and CWE/CVE mapped.
- **API Security Assessment** — REST/GraphQL/SOAP testing against the OWASP API Security Top 10.
- **Secure SDLC Enablement** — threat modeling, secure coding standards, CI/CD security gates (OWASP SAMM/BSIMM aligned).
- **Application Architecture Review** — trust boundaries, data flows, privilege structure (NIST SP 800-53 aligned).
- **Software Composition Analysis (SCA)** — SBOM, CVE findings, license compliance.
- **Malware & Binary Analysis** — tampering and supply chain compromise detection in compiled packages.

## AI Security Across the Platform

- **SAST for AI-generated code** — the same interprocedural taint analysis that catches human mistakes catches the injection flaws, hardcoded secrets, and missing validation AI coding assistants routinely introduce.
- **DAST for LLM applications** — tests chatbots, LLM-backed APIs, and RAG pipelines against the OWASP Top 10 for LLM Applications (2025), with request/response evidence.
- **MAST for AI in mobile apps** — detects hardcoded AI/LLM API keys, exposed on-device model files, and insecure AI endpoint configuration in APK/AAB/IPA binaries.
- **Offline AI** — the AI reasoning behind Autonomous Red Teaming, the AI Pentester and false-positive triage runs inside the customer's appliance, so no target data reaches a third-party AI provider.
- **Regulatory readiness** — EU AI Act (Article 15 robustness & cybersecurity; penalties to EUR 35M or 7% of turnover), Cyber Resilience Act 24-hour vulnerability reporting (from 11 Sep 2026), NIS2 secure development duties, ISO/IEC 42001 — delivered by an ISO/IEC 27001-certified organization.

## What's New (2026)

Full dated log: https://offensive360.com/changelog/

- **2026-09-07 · DAST** — access-control testing hardened: 18-field mass-assignment dictionary, price/quantity tampering confirmed against server-computed outcomes, multi-identity IDOR/BOLA + BFLA authorization differ wired into the engagement engine, header-check false-positive class removed, dashboard splits Scan Risk Index vs External Exposure Risk.
- **2026-09-06 · Red Teaming / ASM / Appliance** — AI Pentester and AI Red Teaming are separate workspaces; ASM monitors accept URLs and names; Azure VHD appliance validated end to end (SAST+DAST+MAST, online and air-gapped).
- **2026-09-05 · ASM** — organization-level risk roll-up; tighter egress controls; AI-assisted triage.
- **2026-09-04 · DAST / Red Teaming** — expanded subdomain enumeration, active port discovery, AI-assisted false-positive triage, offline reasoning fallback.
- **2026-08-31 · SAST / Appliance** — full offline analysis parity for every core language engine; Autonomous Red Teaming engine with scope guard, safe mode and kill switch; first-boot self-heal.
- **2026-08-20 · Appliance** — encrypted (LUKS) Azure payload with relay key release; syslog forwarding to SIEM validated with FortiSIEM.
- **2026-08-02 · SAST** — Free for Open Source program, GitHub Action (SARIF), GitLab CI template, MCP server.
- **2026-08-01 · MAST** — new mobile engine, AI-assisted false-positive triage, detailed PDF report, chunked upload.
- **2026-07-13 · AI Pentester** — authorization-gated PTES engagements with signed authorization, kill switch and OWASP WSTG + MITRE ATT&CK reporting.

## Comparison vs Competitors

| | Offensive360 | Veracode | Checkmarx | Snyk | SonarQube |
|---|---|---|---|---|---|
| SAST | Yes | Yes | Yes | Yes | Yes |
| MAST (mobile) | Yes (built-in) | Add-on | Add-on | No | No |
| DAST | Yes (built-in) | Add-on | Add-on | No | No |
| Attack Surface Management (ASM) | Yes (built-in) | No | No | No | No |
| Autonomous Red Teaming | Yes (built-in, scope-guarded) | No | No | No | No |
| AI Pentester (authorization-gated) | Yes | No | No | No | No |
| SCA | Yes (built-in) | No | Add-on | Yes | Add-on |
| Malware analysis | Yes (unique) | No | No | No | No |
| License compliance | Yes (built-in) | No | Add-on | Add-on | No |
| On-premise appliance (OVA / Azure VHD) | Yes | No | Yes (complex) | No | Yes |
| Air-gapped / offline (incl. AI) | Yes | No | No | No | No |
| Languages built-in | 60+ | 30+ | 30+ | 20+ | 30+ |
| Code upload required | No | Yes | Optional | Yes | Optional |

## Contact & Links

- Website: https://offensive360.com
- SAST product: https://offensive360.com/products/sast/
- DAST product: https://offensive360.com/products/dast/
- ASM (Attack Surface Management): https://offensive360.com/products/asm/
- Autonomous Red Teaming: https://offensive360.com/products/autonomous-red-teaming/
- AI Pentester: https://offensive360.com/products/ai-pentester/
- MAST product: https://offensive360.com/products/mast/
- Product updates / release notes: https://offensive360.com/changelog/
- Security services (incl. AI cybersecurity): https://offensive360.com/services/
- Platform overview: https://offensive360.com/platform/
- Integrations: https://offensive360.com/integrations/
- Pricing: https://offensive360.com/pricing/
- Book a demo: https://offensive360.com/demo/
- Free for Open Source: https://offensive360.com/free-for-open-source/
- Knowledge Base: https://offensive360.com/knowledge-base/
- Compare: https://offensive360.com/compare/
- In-depth guides: https://offensive360.com/blog/best-sast-tools-air-gapped-environments/ · https://offensive360.com/blog/dast-for-llm-applications-owasp-llm-top-10/ · https://offensive360.com/blog/mast-tools-compared-owasp-mobile-top-10/ · https://offensive360.com/blog/sca-plus-malware-analysis-supply-chain-gap/
- GitHub: https://github.com/offensive360
- Security contact: https://offensive360.com/.well-known/security.txt

## About

Offensive360 was founded by security researchers who experienced the gap between what SAST tools promised and what they actually delivered. The platform is built with current security research, continuously updated, and designed for organizations that need real security coverage — not checkbox compliance.

Legal entities: O360 B.V. (Amsterdam, Netherlands; KVK 88260585) and Offensive 360 LLC (UAE).
Headquarters: Netherlands
Founded: 2017
Certifications: ISO/IEC 27001
Target customers: Enterprise security teams, government and defense, financial services, healthcare, software development organizations

---

# Extended reference (llms-full.txt)

This file extends https://offensive360.com/llms.txt with the complete FAQ of every product page and the dated product-updates log. Product pages remain the canonical source.

## SAST — Static Application Security Testing — FAQ

Source: https://offensive360.com/products/sast/

**Q: What programming languages does Offensive360 SAST support?**
Offensive360 SAST supports 60+ programming languages including Java, C#, Python, PHP, JavaScript, TypeScript, Go, Ruby, Kotlin, Swift, Objective-C, Dart, C, C++, Scala, Rust, Groovy, VB.NET, COBOL, Apex (Salesforce), Oracle Forms, ColdFusion, Perl, Elixir, Lua, and more. All language engines are built-in — no add-ons or plugins required.

**Q: Is Offensive360 SAST free for open-source projects?**
Yes. Public repositories on GitHub, GitLab, Bitbucket, or Codeberg can request a free scan token and run the full SAST engine from CI — including the offensive360/sast-scan-action GitHub Action with SARIF output for the code scanning tab. Tokens are valid 30 days and renewable on request. See offensive360.com/free-for-open-source for details.

**Q: How do developers and AI coding assistants use Offensive360 SAST?**
Wherever code is written or merged. IDE plugins are available for VS Code, Visual Studio, IntelliJ, Android Studio and Eclipse. In CI, the offensive360/sast-scan-action GitHub Action runs a full scan and uploads SARIF to the code-scanning tab, and a GitLab CI template is included; Bitbucket, Azure DevOps, Jenkins and CircleCI are supported through the REST API and a built-in setup wizard. For AI coding assistants, the open-source Offensive360 MCP (Model Context Protocol) server lets any MCP-capable assistant scan a local project and read back findings with file, line, severity and a fix — so an assistant can scan, read the criticals, and start patching.

**Q: Can Offensive360 SAST find vulnerabilities in AI-generated code?**
Yes — and it matters more every year. Independent 2025 research across 100+ AI models found roughly 45% of AI-generated code contains known security flaws, with Java worst-affected, and 2026 follow-up testing shows security quality has stalled even as fluency improves. Offensive360 SAST analyzes data flow, not authorship: the same interprocedural taint analysis that catches human mistakes catches the injection flaws, hardcoded secrets, and missing input validation that AI assistants routinely introduce — before they merge.

**Q: Can Offensive360 SAST run offline or in an air-gapped environment?**
Yes. Offensive360 is unique in offering 100% offline operation with zero internet dependency. It deploys as a virtual appliance — an OVA for your data center or an Azure VHD image for your own cloud tenant — and runs in any on-premise or air-gapped environment, including classified networks and defense infrastructure. Every language engine analyzes fully offline; no cloud connection is ever required.

**Q: How does Offensive360 SAST compare to Veracode, Checkmarx, or Snyk?**
Offensive360 differs from Veracode, Checkmarx, and Snyk in several key ways: (1) It runs fully on-premise with no code upload to external servers; (2) It supports 60+ languages vs ~30 for competitors; (3) It includes DAST, SCA, malware analysis, and license compliance in one platform at no extra cost; (4) It is the only commercial SAST/DAST platform with built-in malware and binary analysis; (5) It offers 100% air-gapped operation, which cloud-only tools like Veracode cannot provide.

**Q: Does Offensive360 provide remediation guidance, or just find vulnerabilities?**
Offensive360 finds AND fixes vulnerabilities. Every finding includes the complete data-flow trace from source to sink, a secure code example in your specific language and framework, and step-by-step remediation guidance. Developers can resolve issues without security expertise or guessing.


## DAST — Dynamic Application Security Testing — FAQ

Source: https://offensive360.com/products/dast/

**Q: What is the difference between DAST and SAST?**
SAST (Static Application Security Testing) analyzes your source code without running it — it finds vulnerabilities early, at the code level. DAST (Dynamic Application Security Testing) tests your running application from the outside, the way a real attacker would, so it finds runtime issues like server misconfigurations, authentication flaws, and exploitable injection points that only appear in a live environment. Offensive360 includes both in one platform, so you cover code and runtime without buying two tools.

**Q: What vulnerabilities does Offensive360 DAST test for?**
The engine runs more than 40 active exploit checks and 19 passive analyzers on every scan. Active checks cover SQL injection (error, boolean and time-based), reflected, stored, DOM-based and blind XSS, SSRF, XXE, OS command injection, path traversal, server-side and client-side template injection, SSI, LDAP, XPath, NoSQL and expression-language injection, CRLF and email-header injection, host-header injection, HTTP parameter pollution, prototype pollution, insecure deserialization, HTTP request smuggling, web cache poisoning, CSRF, JWT misconfiguration and replay, session fixation, unrestricted file upload, exposed files and source code, default credentials, password-reset poisoning, open redirects, weak TLS, GraphQL introspection, IDOR/BOLA, missing function-level authorization (BFLA), mass assignment, and API inventory and resource-consumption weaknesses. Passive analyzers cover CORS, security headers, cookie flags, clickjacking, secrets in responses and URLs, version disclosure, cache control, subdomain takeover, anti-CSRF tokens, error leakage, directory listing, mixed content, exposed source maps, session-token entropy, missing login rate limiting, form security, and vulnerable or outdated components. Fingerprint packs add known-CVE matching for WordPress, Joomla, Drupal and common frameworks.

**Q: Does DAST work behind login pages?**
Yes. Offensive360 DAST supports authenticated scanning: form-based login, token and cookie sessions, multi-step login flows, and role-based testing across privilege levels. Since most of an application’s attack surface sits behind authentication, this typically raises coverage from roughly 20% to 90% of the application.

**Q: Can Offensive360 DAST test single-page applications and GraphQL APIs?**
Yes. A headless-Chromium crawler renders JavaScript-heavy single-page applications to discover the routes, forms and API calls a plain HTTP spider never sees. GraphQL endpoints are checked for introspection exposure, and both REST and GraphQL APIs are tested for injection, broken object-level authorization, mass assignment and missing resource-consumption controls. Blind and out-of-band vulnerabilities — SSRF, XXE, blind XSS — are confirmed through a self-hosted out-of-band (OAST) callback service, so no third-party collaborator server is involved.

**Q: Can Offensive360 DAST test AI and LLM-powered applications?**
Yes. Applications with chatbots, LLM-backed APIs, or RAG features have a new attack surface that traditional scanners miss. Offensive360 tests AI-powered applications against the OWASP Top 10 for LLM Applications (2025) — prompt injection (the #1 LLM risk), jailbreaks, system prompt leakage, improper output handling, and sensitive information disclosure — alongside the classic web and API vulnerability classes.

**Q: Are DAST findings validated, or will I get false positives?**
Findings are validated with real exploit payloads and reported with the request/response evidence that proves them — you see exactly what was sent and what came back. Post-scan validators re-check every finding, an AI-assisted triage pass flags anything that looks like a false positive, and every engine release must produce zero findings against clean control applications before it ships. Safe mode is on by default, so validation never damages the target.

**Q: How does DAST relate to the AI Pentester, Autonomous Red Teaming, and ASM?**
They share one engine. A DAST scan is the point-in-time, deterministic test of a target. The AI Pentester runs a full, authorization-gated engagement on top of it, with human approval before exploitation. Autonomous Red Teaming runs continuously inside an enforced scope guard and proves exploitability with non-destructive proof-of-exploit. Attack Surface Management (ASM) discovers the external assets — subdomains, services, certificates, leaked credentials — that can then be handed to any of the three for authorized testing. All of it lives in one dashboard, which separates the findings-weighted Scan Risk Index from the ASM-derived External Exposure Risk.

**Q: Can Offensive360 DAST run on-premise or air-gapped?**
Yes. DAST ships inside the same virtual appliance as SAST, MAST and ASM — an OVA for your data center or an Azure VHD image for your own cloud tenant — and runs fully offline in air-gapped networks. Targets, findings and evidence never leave your environment.


## MAST — Mobile Application Security Testing — FAQ

Source: https://offensive360.com/products/mast/

**Q: What is MAST (Mobile Application Security Testing)?**
MAST analyzes mobile application binaries and source for security vulnerabilities. Offensive360 MAST inspects Android APK/AAB and iOS IPA packages — the manifest, permissions, code, and embedded resources — and maps every finding to the OWASP Mobile Top 10 (2024).

**Q: Do I need source code to scan a mobile app?**
No. Offensive360 MAST works directly on the compiled binary — upload the APK, AAB, or IPA you ship to the store and get results in minutes. If you also scan the source with Offensive360 SAST (Kotlin, Java, Swift, Dart and more are supported), you get full-stack mobile coverage.

**Q: Which vulnerabilities does MAST detect?**
Hardcoded secrets and API keys, insecure data storage, cleartext traffic and ATS misconfigurations, debuggable release builds, dangerous or exported components, weak cryptography, excessive permissions, missing binary protections, and vulnerable embedded SDKs — each classified against the OWASP Mobile Top 10 (2024).

**Q: Does MAST detect AI-related risks in mobile apps?**
Yes. Mobile apps increasingly embed AI: LLM API calls, on-device models, and AI SDKs. MAST detects the risks that come with them — hardcoded AI/LLM API keys in the binary (2025 mobile-threat research found nearly half of apps contain hardcoded secrets, and production apps have been caught leaking live LLM API keys), exposed model files that can be extracted from unprotected packages, and insecure communication with AI endpoints.

**Q: What does a MAST report include?**
Every scan produces a detailed PDF report: each finding is mapped to its OWASP Mobile Top 10 (2024) category with severity, the evidence found in the binary, and remediation guidance. An AI-assisted triage pass marks likely false positives before you read it, so reviewers spend their time on real risk. Findings also appear in the same console as your SAST and DAST results.

**Q: Can MAST run on-premises or air-gapped?**
Yes. MAST ships inside the same virtual appliance as Offensive360 SAST and DAST — an OVA for your data center or an Azure VHD image for your own cloud tenant — and runs fully offline. Your mobile binaries never leave your network — ideal for government, defense, banking, and healthcare teams.


## ASM — Attack Surface Management — FAQ

Source: https://offensive360.com/products/asm/

**Q: What is Attack Surface Management (ASM)?**
Attack Surface Management is the continuous discovery, inventory, and monitoring of everything your organization exposes to the internet — domains, subdomains, hosts, open ports, services, certificates, and leaked credentials — so you find and fix exposures before an attacker reaches them. Offensive360 ASM runs this discovery continuously from a seed scope, scores each exposure by risk, and alerts you when your external surface changes.

**Q: How does Offensive360 ASM discover assets?**
From a seed domain (or a list of domains and emails), the engine enumerates subdomains through passive sources and active brute-forcing, resolves hosts, discovers open ports and running services, inspects TLS/SSL certificate posture, and correlates OSINT signals such as leaked emails and breached credentials. Every discovered asset is de-duplicated, scoped to your organization, and tagged with the exposures found on it.

**Q: Does ASM monitor my attack surface continuously?**
Yes. You define one or more monitors over a scope of domains and alert emails, choose an hourly or daily cadence, and the engine re-runs discovery on schedule. When a new asset, port, service, or exposure appears — or a breached credential is found — it is recorded against your organization and can raise an alert, so a newly exposed host or an expiring certificate does not sit unnoticed.

**Q: Is Offensive360 ASM safe and non-intrusive?**
Yes. Discovery is designed to be safe against production: it is bounded to the scope you authorize, its outbound reconnaissance runs behind an egress control so it cannot wander outside your assets, and it does not exploit or disrupt the services it finds. ASM builds the exposure picture; the Offensive360 DAST engine and AI Pentester handle any authorized, deeper testing.

**Q: Can ASM run on-premise or air-gapped?**
Yes. Offensive360 ASM deploys via the Offensive360 OVA and runs fully on-premise. In air-gapped networks it discovers and monitors the internal surface you scope, with no requirement to send asset data to a cloud service — asset inventory and exposure records never leave your environment.

**Q: Does ASM detect breached and leaked credentials?**
Yes. During discovery the engine correlates OSINT breach sources for the domains and emails in scope, surfacing leaked or breached credentials tied to your organization so you can force resets and close the exposure. Breach findings are scored and prioritized alongside the rest of your external surface.

**Q: How is Offensive360 ASM different from a standalone ASM vendor?**
Standalone ASM products are typically cloud-only and stop at an inventory. Offensive360 ASM is part of one unified platform: the assets and exposures it discovers feed directly into the DAST engine and the AI Pentester for authorized validation, it deploys on-premise and air-gapped, and its risk scoring prioritizes what to test next — discovery, validation, and reporting in one appliance rather than a bolt-on.


## Autonomous Red Teaming — FAQ

Source: https://offensive360.com/products/autonomous-red-teaming/

**Q: What is autonomous red teaming?**
Autonomous red teaming is continuous, machine-speed adversary emulation: an engine that plans and runs its own attack paths against your authorized scope — reconnaissance, chaining findings into attack paths, and validating exploitability — instead of a human driving every step. Offensive360 runs this autonomously inside an enforced authorization boundary, defaults to non-destructive proof-of-exploit, and can be stopped instantly with a kill switch.

**Q: Is “fully autonomous” red teaming safe to run?**
It is safe because the autonomy is bounded. The engine plans and validates on its own, but an enforced scope guard keeps it inside the assets you authorize, denial-of-service techniques are force-disabled, safe-mode (non-destructive proof-of-exploit) is the default, and a visible kill switch halts a running operation instantly. Autonomy applies to the planning and validation work — not to leaving your boundary or damaging your systems.

**Q: What is the scope guard?**
The scope guard is an enforced boundary around every autonomous operation. Before the engine acts on a target it checks that the target is inside the scope you authorized; anything out of scope is refused. It is the mechanism that lets red teaming run autonomously and continuously without wandering onto assets you did not approve — the enforced difference between an autonomous red team and an unleashed one.

**Q: How is this different from the Offensive360 AI Pentester?**
They are two modes of the same platform. The AI Pentester runs a point-in-time, PTES-methodology engagement gated by a signed authorization record with human approval before exploitation — built for formal, audit-ready pentests. Autonomous Red Teaming runs continuously and autonomously inside the enforced scope guard, chaining attack paths and proving exploitability at machine speed. Many teams run both: continuous autonomous coverage, plus a formal human-approved engagement when they need one.

**Q: Does it actually exploit, or just report potential issues?**
It validates. Rather than reporting a long list of "possible" issues, the engine attempts a safe, non-destructive proof-of-exploit and marks a finding confirmed only when it demonstrably fires — capturing the request, the response, and a reproducible proof. That is how autonomous red teaming cuts false positives: a confirmed finding comes with evidence you can replay, not just a severity label.

**Q: Can it run on-premise, air-gapped, and offline?**
Yes. Autonomous Red Teaming deploys via the Offensive360 OVA and runs fully on-premise. Its AI reasoning can run offline inside an air-gapped network, so targets, attack paths, and proofs never leave your environment — a capability autonomy-first cloud tools generally cannot offer.

**Q: How is it different from tools like XBOW, NodeZero, Pentera, or Horizon3?**
Autonomy-first tools in that category are generally cloud-delivered and lead with running end-to-end at machine speed. Offensive360 is distinctive in combining autonomous, scope-guard-enforced operation with proof-of-exploit validation, offline and air-gapped AI, a visible instant kill switch, and a complementary human-approved engagement mode (the AI Pentester) — autonomy inside an enforced boundary, deployed where your most sensitive targets actually live.


## AI Pentester (authorization-gated) — FAQ

Source: https://offensive360.com/products/ai-pentester/

**Q: What is an AI pentester?**
An AI pentester is an AI-driven capability that plans and runs a penetration-testing engagement — reconnaissance, vulnerability analysis, exploitation, and reporting — instead of a human consultant driving every step by hand. The Offensive360 AI Pentester runs authorized engagements following the PTES methodology inside the Offensive360 platform, and only proceeds after a signed authorization record is in place, with a human approving before any exploitation.

**Q: Is AI penetration testing safe and authorized?**
Yes, when it is governed. The Offensive360 AI Pentester requires a first-class, in-product authorization record before any engagement runs: a named signatory, eight explicit attestations, and a typed signature that must exactly match the signatory name, captured with the signer’s IP and timestamp into an immutable record — and the launch is blocked server-side on any mismatch. A human must approve before exploitation, denial-of-service techniques are force-disabled, and a visible kill switch stops the engagement instantly.

**Q: Does the AI Pentester run fully autonomous, unsupervised exploitation?**
No. Exploitation is paused until a human explicitly approves it, denial-of-service techniques are force-disabled and cannot be enabled, and a visible one-click kill switch stops the engagement instantly at any time. The AI drives the reconnaissance, analysis and reporting work at machine speed, but the moment of exploitation stays under human control.

**Q: Can I stop an AI pentest engagement while it is running?**
Yes. A visible one-click kill switch lets the operator halt a running engagement instantly at any moment. Combined with the human-approval pause before exploitation and the force-disabled DoS techniques, your team keeps a hard stop on the engagement from start to finish.

**Q: Is the Offensive360 AI Pentester an autonomous pentest tool like XBOW or NodeZero?**
Tools in the autonomous-pentest category — such as XBOW, Horizon3.ai NodeZero, Pentera, RidgeBot, and Terra Security — are generally autonomy-first, designed to run end-to-end at machine speed. The Offensive360 AI Pentester is control-first. Among these tools it is distinctive in combining a first-class signed-authorization gate, a visible instant kill switch, a default human-approval pause before exploitation, and true on-premise and air-gapped deployment — authorized, human-in-the-loop engagements rather than autonomy for its own sake.

**Q: What is the best AI penetration testing tool for on-premise or air-gapped environments?**
For teams that cannot send engagement data to a cloud service, the Offensive360 AI Pentester is purpose-built: it deploys via the Offensive360 OVA and runs fully on-premise and in air-gapped networks, so target data and engagement records never leave the customer network. That makes it suitable for government, defense, finance, and critical-infrastructure buyers who need authorized AI pentesting with in-network proof of authorization.

**Q: Can the AI Pentester run on-premises or air-gapped?**
Yes. The AI Pentester deploys via the Offensive360 OVA and runs fully on-premise and in air-gapped networks, so target data and engagement records never leave the customer network. There is no requirement to send engagement data outside your environment.

**Q: What methodology and standards does the AI Pentester follow?**
It runs authorized engagements in explicit PTES phases: Pre-engagement (scope, Rules of Engagement, and the signed authorization record), Reconnaissance with OSINT intelligence insights, Vulnerability Analysis via a real Offensive360 DAST scan, human-approved Exploitation, and Reporting. Every finding in the AI-generated report is mapped to both the OWASP Web Security Testing Guide (WSTG) and MITRE ATT&CK, with prioritized severity, remediation guidance, and remediation SLAs (target timelines).


## Product updates log

Source: https://offensive360.com/changelog/ (newest first)

### 2026-09-08 — Website: Product updates page and full DAST coverage documentation
- This product-updates page is live and linked from every page footer.
- The DAST product page now lists every active and passive check class the engine runs.
- llms.txt and the new llms-full.txt describe all six products, their FAQs and this update log for AI assistants and search engines.

### 2026-09-07 — DAST: Access-control and business-logic testing hardened
- Mass-assignment checks now probe an 18-field dictionary of privileged attributes (role, isAdmin, verified, balance, owner_id, tenant_id, price and more) in form and JSON bodies — up from 3.
- Price and quantity tampering in AI-driven engagements is confirmed against the server-computed outcome, not an echoed value.
- A multi-identity authorization differ — IDOR/BOLA and function-level BFLA across high, low and anonymous identities — is wired into the engagement engine.
- Security-header and clickjacking checks no longer report on responses that were never captured, removing a class of false positives.
- The Vulnerabilities view shows human-readable finding types, and the CVSS column appears only when scores exist.
- The dashboard separates the findings-weighted Scan Risk Index from the ASM-derived External Exposure Risk, and its counts match the Scans list.

### 2026-09-06 — DAST · Red Teaming · ASM · Appliance: AI Pentester and AI Red Teaming become separate workspaces
- AI Red Teaming operations get their own sidebar tab, list and creation flow, distinct from approval-gated AI Pentester engagements.
- ASM monitor creation accepts full URLs (reduced to hostnames) and named monitors.
- Azure VHD appliance images validated end to end alongside the OVA — SAST, DAST and MAST, online and fully air-gapped.

### 2026-09-05 — ASM: Organization risk roll-up and tighter egress controls
- Per-exposure risk scores roll up into an organization-level exposure score on the dashboard.
- Tighter egress controls for OSINT collection, and an AI-assisted triage pass that marks likely false positives.

### 2026-09-04 — DAST · Red Teaming: Wider discovery and offline fallback
- Subdomain enumeration expanded and active port discovery added ahead of crawling.
- AI-assisted false-positive triage on scan findings.
- Red-team operations fall back to offline reasoning automatically when no cloud AI service is reachable.

### 2026-08-31 — SAST · Red Teaming · Appliance: Full offline analysis parity on air-gapped appliances
- All core language engines now analyze fully offline — verified by scanning the same vulnerable code with the network disconnected.
- The Autonomous Red Teaming engine ships with an enforced scope guard, safe mode by default, denial-of-service force-disabled and a kill switch.
- The appliance engine self-heals after an interrupted first boot.

### 2026-08-20 — Appliance: Encrypted Azure payload and SIEM forwarding
- Azure images carry an encrypted (LUKS) payload with key release through the licensing relay, so a cloud tenant cannot read the engine at rest.
- Syslog forwarding to SIEM platforms validated end to end with FortiSIEM.

### 2026-08-02 — SAST: Free for Open Source, GitHub Action and MCP server
- Public repositories on GitHub, GitLab, Bitbucket and Codeberg can request a free 30-day SAST scan token.
- offensive360/sast-scan-action runs a full scan in GitHub Actions and uploads SARIF to the code-scanning tab; a GitLab CI template is included.
- An open-source MCP server lets AI coding assistants scan a local directory and read findings with file, line, severity and fix.

### 2026-08-01 — MAST: Mobile engine overhaul
- New mobile analysis engine for Android APK/AAB and iOS IPA binaries.
- AI-assisted false-positive triage and a detailed PDF report mapped to the OWASP Mobile Top 10 (2024).
- Chunked upload for large binaries.

### 2026-07-13 — AI Pentester · DAST: Authorization-gated AI penetration testing
- PTES-structured engagements on the DAST engine: reconnaissance, vulnerability analysis, human-approved exploitation and reporting.
- Signed in-product authorization record with eight attestations, a visible kill switch and denial-of-service force-disabled.
- OWASP WSTG and MITRE ATT&CK mapped reports.
- LLM application probes for the OWASP Top 10 for LLM Applications (2025).