Skip to main content
TL;DR: Vijil is the trust layer for AI agents, measuring reliability, security, and safety before deployment (Diamond), protecting Agents at runtime (Dome), and evolving them through real-world feedback (Darwin). Start with Concepts for the mental model, or jump to the Owner’s Guide or Developer Guide to get started. If you do not yet know what is running in your estate, begin with Discover.
Agents often stall before production: security and compliance need proof, not demos. Vijil turns that into measurable trust, systematic evaluation, runtime protection, and a feedback loop from real behavior. Trust Score summarizes how an Agent performs across reliability, security, and safety, giving reviews evidence instead of opinions. Diamond runs Trust Score evaluations and adaptive Red Team evaluations. Dome enforces Guardrails at runtime, while Darwin uses reinforcement learning over production telemetry to continuously improve an Agent’s resilience. Before any of that, Discover finds the AI agents, models, and services actually running across your estate, including the ones nobody registered, and hands the ones worth governing to the Agent Registry.

​
How Vijil Measures Trust

Start with the full model in Trust Score.

​
Build, Ship, Run, Evolve

Agent prototypes are everywhere, but most do not make it to production. To scale AI, enterprises face a critical choice: ship fast or ship resilient. Teams that try to do both run into the following challenges:
  • Shipping fast increases failure rates and elevates risk; without objective trust evidence, Agents are in security limbo
  • Generic red-teaming and Guardrails do not ensure Agents are safe for their specific context
  • Failure insights do not feed back into development, leaving the trust gap open
Vijil closes that gap at each stage of the Agent lifecycle:

Discover

Finds the AI running in your estate and classifies it, so the inventory you evaluate against is what is actually deployed rather than what teams remembered to mention.

Diamond

Automates evaluation based on custom policies for business context and personas. Turns “is this safe?” into an answerable question with an audit-ready Trust Score.

Dome

Enforces policies to protect and monitor Agents at runtime through embedded Guardrails, blocking attacks in real time and logging detailed telemetry.

Darwin

Learns from production telemetry and feeds those lessons back into the Agent to improve its resilience over time. Darwin is in development and not yet generally available.

​
Continue with Vijil

Pick where you work and what you need next:

Concepts

You want the mental model - Trust Score, evaluation components (Harness, Scenario, Probe), and Defense (Guards, Guardrails, Detectors).Start here → Trust Score, then explore this tab for deeper topics and the Glossary.

Agent Owner's Guide

You use the Vijil console to register Agents, run Evaluations, configure Dome, monitor Dome Metrics, and report to stakeholders.Start here → Get started in the console.

Agent Developer's Guide

You integrate Vijil via SDKs & APIs, frameworks (e.g. LangChain, ADK), and production wiring.Start here → Developer introduction and installation.

​
Quick Resources

Last modified on November 4, 2026