Security policies and bindings
Bundle checks into policies and bind them to everyone, a group, a model, a provider or a service token.
What it is
A security policy is a named bundle of checks bound to the organization, a group, a service token, an upstream, a model, an alias or personal subscriptions. Each check runs in observe, redact or block mode on requests, responses or both.
Why you want it
Different traffic needs different rules. Binding policies separately from grants means 'block secrets everywhere' is one setting, with no service-token bypass.
How it works
- Check families: request shape, secrets, PII, term lists, prompt injection, content safety
- Fail open or fail closed per check
- Violations recorded with action taken and retained per kind
- Overview, policies, bindings, term lists, violations and classifiers tabs
In the product
A mandatory company baseline that redacts secrets and personal data and blocks codenames, plus a stricter policy for a customer-facing bot.
Actual Janus interface · Demo identities and synthetic usage
What fired in the last 24 hours, by check, and the steps from observing to enforcing.
Actual Janus interface · Demo identities and synthetic usageSee it on your own network.
The Community edition is free for up to 25 people. The 30-day Business trial unlocks every Business feature.