Agents API
An Agent is the core identity unit in Synentra. Every autonomous AI agent, service, or system that proxies requests through Synentra must be registered as an agent. Each agent has a unique identity, a trust score, an assigned policy, and a lifecycle status.
Policies API
A Policy defines the authorization and governance rules that determine what an AI agent is permitted to do. Policies consist of prioritized rules, each containing one or more conditions that evaluate incoming requests.
Tokens API
Exchange an agent's ID and credentials for a JWT access token.
Proxy API
Synentra acts as a reverse proxy gateway. Any HTTP request prefixed with /proxy/ is intercepted, evaluated, and either forwarded to the upstream service, blocked, or put on hold for human-in-the-loop (HITL) approval.
Health API
Synentra exposes a health check endpoint that can be used for liveness and readiness probes, as well as for monitoring the overall health of the gateway and its dependencies.
HITL API
The Human-in-the-Loop (HITL) system allows operators to manually review and approve or deny high-risk requests that have been intercepted by the decision engine. This ensures human oversight for sensitive operations, such as bulk deletes or actions that exceed a risk threshold.