EvalOps builds control-plane software for AI agent workforces. We help teams see which agents, assistants, and tools are taking action, govern what each one is allowed to do, and widen autonomy only when the evidence supports it.
- Visibility across AI agents, assistants, tools, and the actions they take.
- Policy, approval, evaluation, cost, and audit rails for agentic systems.
- Developer workflows that turn agent output into reviewable, repeatable work.
- Governance patterns for the AI fleet companies already have in production.
Discover repos by topic:
evalops-platform- core services and contractsevalops-product- product frontends and agent experiencesevalops-infra- deployment and infrastructureevalops-data- data pipelines and analyticsevalops-docs- documentation
| Repo | Description |
|---|---|
| platform | Control-plane monorepo for services, contracts, and console UI |
| maestro | Multi-model coding agent (TUI, IDE, Slack, GitHub) |
| deploy | GitOps delivery: K8s manifests, Terraform, ArgoCD |
| hopper | Marketing site |
Report vulnerabilities through SECURITY.md. For general contact, use the project-specific repository or the contact path published on the EvalOps website.