NOVASECLABS / KNOWLEDGE BASE
Understand the system.
Interrogate the evidence.
Product context, technical principles, application-security practice, and enterprise evaluation resources for VulnGPT.
DOCUMENT LIBRARY
Find what you need.
Search by topic or narrow the library by audience.
VulnGPT is in private beta. Documentation describes product direction and evaluation guidance; it does not guarantee current feature availability.
Showing all documents
Product overview
What VulnGPT is being built to investigate, who it serves, and where its boundaries are.
Read document ↗ TECHNICAL FOUNDATION 02Technical approach
A conceptual model for investigating behavior across identity, authorization, and application state.
Read document ↗ SYSTEM MODEL 03System overview
Public conceptual architecture, trust boundaries, and the separation between test context and evidence review.
Read document ↗ EVIDENCE STANDARD 04Evidence and validation
How to distinguish a suspicious signal from a supported, reviewable security conclusion.
Read document ↗ PRACTITIONER GUIDE 05Security testing playbook
Controlled patterns for object ownership, roles, multi-step workflows, sessions, and ambiguous results.
Read document ↗ EVALUATION TOOL 06Capability evaluation matrix
Scenario dimensions, required context, evidence expectations, and common false conclusions.
Read document ↗ REPRODUCIBLE METHOD 07Evaluation protocol
Define ground truth and outcomes before a run; separate execution, observability, correctness, and evidence quality.
Read document ↗ REFERENCE 08Application security use cases
Representative application behaviors that motivate contextual security testing.
Read document ↗ REFERENCE 09Security glossary
Shared definitions for application-security, evidence, authorization, and workflow terminology.
Read document ↗ BUYER GUIDE 10Enterprise evaluation
A structured diligence guide for security, architecture, procurement, privacy, and product stakeholders.
Read document ↗ TRUST CENTER 11Security and privacy
Public trust boundaries and the specific questions that require current vendor confirmation.
Read document ↗ TRUST MODEL 12Security model
Conceptual actors, data assets, trust boundaries, and threat scenarios for evaluation.
Read document ↗ WORKSHEET 13Assessment plan
Define scope, identities, operational constraints, data handling, approvals, and success criteria.
Read document ↗ REPORTING STANDARD 14Finding report standard
Structure a finding around the expected boundary, evidence, impact, uncertainty, and retest.
Read document ↗ PUBLICATION POLICY 15Benchmark governance
Reproducibility, denominator integrity, independent review, and benchmark publication gates.
Read document ↗ CLAIM CONTROL 16Claims and messaging
Approved product language and guardrails for performance, security, and assurance statements.
Read document ↗ MAINTAINERS 17Release process
Review, verification, publication, and correction procedures for the public showcase.
Read document ↗ SAFE USE 18Responsible testing
Authorization, scope, data minimization, safe execution, and coordinated disclosure guidance.
Read document ↗ MAINTAINERS 19Repository maintenance
Local validation, asset hygiene, public-content review, and repository ownership.
Read document ↗ QUICK ANSWERS 20Frequently asked questions
Product status, access, intended use, supported claims, and common evaluation questions.
Read document ↗ MACHINE-READABLE RESOURCES 21Public evaluation schemas
Optional JSON formats for externally documenting findings and benchmark reports—not VulnGPT API contracts.
Read schemas ↗No matching documents
Try another phrase or clear the current filters.
IMPORTANT CONTEXT
Public documentation.
Bounded claims.
VulnGPT is in private beta. These documents describe product direction, conceptual methods, and evaluation resources. They do not guarantee feature availability, security certification, benchmark performance, or contractual controls. Confirm the proposed use case and current terms with NovaSecLabs.