Documentation
Build with Dreadnode
AI security platform for automated red teaming, agent development, evaluation, and experiment tracking
- Quickstart
Run your first capability
Auth the CLI, launch a capability, and inspect the trace.
Read - Red Team
Run an AIRT assessment
Launch jailbreak suites against any model from the CLI or SDK.
Read - SDK
Build a custom capability
Package agents, tools, and skills into a reusable capability.
Read - SDK
Write a custom scorer
Author Python scorers and wire them into an evaluation.
Read - TUI
Inspect sessions and traces
Replay runs, browse transcripts, and debug agents in the terminal.
Read - Self-host
Install with Helm
Deploy the platform onto your own Kubernetes cluster.
Read
For agents
Machine-readable docs
Stitched, plain-text exports for AI agents and offline indexing.