Security and risk teams
You need to know what AI is running, enforce a control that genuinely blocks, prove it works against a named framework, and hand someone the evidence.
Start here
Solutions
Same gateway, 3 different first questions
The gateway does 1 thing: it sits between your applications and your AI services and gives you control over what crosses. What you care about first depends on which side of that you own.
Platform and infrastructure
You need it to be boring: 1 process to deploy, a latency budget it respects, blast-radius control when a provider misbehaves, and metrics without writing instrumentation.
Start hereAgent and app builders
You need client auth to configure itself, tools scoped to what the agent actually calls, PII kept out of the provider, and a spend limit that stops a runaway loop.
Start here
Common ground
What everyone gets regardless
These are not tier-gated or role-specific. They are how the gateway works.
44
Capabilities
Nothing held back behind a tier
$0
To run it
No licence key, no metering
100%
On your network
No telemetry, no call home
Opt-in
Every control
Existing proxies keep behaving
Pick a starting point and run it locally
All 3 paths begin the same way: download it, start it, and create 1 proxy.