| layout | col-sidebar |
|---|---|
| title | OWASP ThreatAtlas |
| tags | threat-modeling |
| level | 2 |
| type | tool |
| pitch | Open-source web platform for collaborative threat modeling—diagrams, frameworks, mitigations, and optional AI assistance in one place. |
OWASP ThreatAtlas is an open-source web application for team-based threat modeling. Organizations run structured sessions by inviting developers, DevOps, architects, and security engineers to map systems, record threats and mitigations, and review risk in one place—instead of scattering notes across documents and slides.
The project is Apache 2.0 licensed. The application source ships as a modern stack (FastAPI, PostgreSQL, React) and is designed to be self-hosted so data stays under your control.
Bridge the gap between security frameworks and real architectures: make it easy to draw data flows, attach threats and mitigations from recognized knowledge bases, collaborate with clear ownership and history, and review how risk changes over time.
- Data Flow Diagrams (DFDs) — Interactive diagrams for processes, data stores, external entities, flows, and trust boundaries, with Draw.io / .drawio import and diagram templates for fast onboarding.
- Component Library — Drag pre-built security components into diagrams and use the Component Threat Library to browse, edit, and revert predefined threat/mitigation mappings per framework.
- Threats, mitigations, and risk — Link mitigations to threats; record likelihood, impact, and risk score; see risk-oriented views in analytics and heatmap overlays.
- Knowledge base — Browse and apply content from multiple frameworks (e.g. STRIDE, PASTA, LINDDUN, OWASP references, MITRE-oriented material, CVSS-oriented guidance—see the app and docs for the current catalog).
- Collaboration — Products shared with teammates; RBAC, invitations, live cursor tracking, real-time diagram sync, auto-save, and visibility controls aligned with how teams actually work.
- Approvals and risk acceptance — Formal approval workflows, justification tracking, a dedicated approvals dashboard, and notifications for reviewers and collaborators.
- Notifications and search — In-app bell with unread badges plus global search (⌘K / Ctrl+K) for products, diagrams, KB threats, and mitigations.
- DevOps integration — API Tokens for machine access, JIRA issue creation from threat cards, and a CI/CD Security Gate endpoint for pass/fail checks.
- History — Diagram versioning with comparison, including visibility into threat and mitigation changes—not only canvas edits.
- Comments — Discussion on threats and mitigations for async review.
- Custom frameworks — Define organization-specific methodology and reuse it across diagrams.
- Optional AI assistant — When enabled by an administrator, a conversational AI assistant in the diagram editor can help explore threats and proposals; provider and keys are configured in-app (see deployment and security notes in the docs).
- Open source and self-hosted — Inspect the code, adapt it, and run it in your environment.
- Many frameworks, one workspace — Combine diagramming, a structured threat/mitigation model, and a growing knowledge base instead of maintaining separate spreadsheets and diagrams.
- Built as a product, not a static site — Authentication, teams, persistence, and UI workflows for day-to-day threat modeling—not only reference pages.
- Extensible methodology — Custom frameworks sit alongside built-in catalogs so teams can encode their own standards.
- Practical outputs — Analytics, versioning, and structured data (e.g. diagram export) support reviews, onboarding, and continuous refinement of a threat model.
- ThreatAtlas application (
threatatlas-app/) — Backend, frontend, Docker Compose, and tool documentation.
Run ThreatAtlas locally or in your infrastructure:
Build, test, and submit changes:
Learn the main UI flows (products, diagrams, threats, mitigations, settings):
What changed in each version: