---
title: "EvalGate — AI quality infrastructure"
description: "Evaluate AI systems with traceable evidence, regression coverage, and release gates."
canonical: "https://www.evalgate.com/"
last-updated: "2026-08-24"
---

# EvalGate

EvalGate is the AI control plane for deciding whether an AI change is safe to ship.

Unit tests tell you whether the code still works. EvalGate tells you whether the AI still behaves.

## Start here

- [Human developer quickstart](https://www.evalgate.com/docs/quickstart)
- [Developer hub](https://www.evalgate.com/developers)
- [Canonical coding-agent Skill source](https://github.com/evalgate/skills)
- Install the complete Skill collection with `npx skills add evalgate/skills`.
- [Authentication](https://www.evalgate.com/auth.md)
- [Feature status](https://www.evalgate.com/docs/platform/feature-status)
- [OpenAPI](https://www.evalgate.com/openapi.json)
- [MCP boundaries](https://www.evalgate.com/mcp.md)

Missing required evidence is not a pass. An offline gate proves only the deterministic evidence that actually ran.

## Capabilities already available

- GitHub Actions workflow generation and annotations, plus a documented GitLab CI recipe.
- Signed outbound webhooks, Slack App delivery when configured, and Slack Incoming Webhook fallback.
- OpenTelemetry/OpenInference-compatible OTLP trace ingestion and export.
- Reviewed synthetic and red-team case generation before production incidents.
- Immutable Prompt Hub versions, baseline/candidate experiments, and protected-environment rollback.
- Non-serving shadow comparison for traffic routed through EvalGate Deployable Assets.

Read the [evidence-backed capability map](https://www.evalgate.com/capabilities.md) for exact entry points and current limits.