Experiment
Preview — execution disabledAgentic Security Lab
Explore how autonomous security agents reason, collaborate, verify evidence, and operate under explicit authority boundaries.
This laboratory is a navigable, technical and educational space for agentic cybersecurity. Each scenario documents a real problem — how agents analyse findings, detect manipulation, correlate evidence, verify claims and act only within explicit authority — through an interactive architecture graph and a step-by-step example trace. The goal is to make the mechanisms legible, not to produce a score.
Preview Mode — Execution Disabled
This laboratory is currently available for architectural exploration and educational demonstrations only. Live agent execution and security testing are disabled. Operational access requires prior explicit authorization from RavenRootviel.
- 8
- Scenarios to explore
- 33
- Agent roles represented
- 8
- Documented architectures
- 8
- Example execution flows
Availability: preview · live execution disabled
Explore
Agent Architecture Explorer
A conceptual, teaching view of the general pipeline: how untrusted inputs and tool outputs become hypotheses, how a deterministic verifier promotes only what it can settle, and how enforcement sits outside the reasoning agent. Drag to pan, scroll to zoom, select any component for its role, inputs and outputs. Each scenario below has its own architecture.
Conceptual teaching view, not a map of the private engine. Reasoning proposes; a deterministic verifier and a policy engine — outside the model — decide what becomes evidence and what is allowed to act. PROVEN / NOT PROVEN / UNKNOWN are kept distinct.
Scenarios
Choose a scenario to explore
Each scenario is self-contained: an overview, an interactive architecture, the agents and their responsibilities, the workflow, a navigable example trace, the evidence and security controls involved, the failure modes, and its limitations.
Multi-Agent Vulnerability Analysis
A pipeline of narrow agents turns raw tool findings into checked evidence — and keeps everything unverified marked UNKNOWN.
5 agents · 8 components
Prompt Injection Detection
Deciding whether text an agent read is data — or an instruction it must refuse.
3 agents · 7 components
RAG Security Assessment
Treating every retrieved document as untrusted data, so a poisoned source cannot steer the answer.
4 agents · 9 components
Multi-Agent Attack Chain Analysis
Several narrow agents stitch independent findings into a candidate attack chain — which stays a hypothesis until independent evidence corroborates it.
5 agents · 9 components
Epistemic Verification
A claim is promoted only when a deterministic, LLM-free check settles it — a model’s confidence is never proof.
4 agents · 8 components
Agent Authorization & Policy Enforcement
An agent can propose any action; a deterministic boundary decides what is allowed to happen.
4 agents · 7 components
Agentic System Failure Modes
When agents in a distributed system die, stall, retry or come back late, correctness has to survive it — not assume it away.
4 agents · 8 components
Security Evidence & Provenance
Tracing how a security record was produced and showing it was not altered — never that its conclusion is true.
4 agents · 7 components
Authorized Access Only — Live Security Testing
Live security testing and agent execution require prior explicit authorization from RavenRootviel. The current preview provides educational content and non-operational demonstrations only. Browsing the educational content is public; operational use of the labs will require express authorization when it is enabled in a later phase. Access & use policy.