Assessment Intelligence
A draft red-team assessment from your own context
- Reports, knowledge-base text, source files and text-based PDFs
- Prioritized scenarios, source references and legitimate controls
- Markdown and JSON exports. Planning does not execute tests.
IntentScan
Adversarial probes against your AI endpoint
- Role transformation, gradual drift and language variation strategies
- Violation analysis with a 0–100 risk score
- Scan history and PDF reports
IntentEnforce
Runtime intent policy in front of your model
- Per-request intent classification
- Allow, block or clarify policy rules
- Proxy layers with request logs
Fair-use limits
Per account, to keep the service free for everyone
- 1,000 tests per month
- 10 requests per minute
- Your model key is encrypted at rest and removable at any time
The one paid service is a hands-on red-team assessment, because it takes our time. It is scoped and quoted per engagement.