DocsBook a demo

Platform

Platform overviewArchitectureWorkflow orchestrationGitOps configurationGovernance and AAAAI and MCPKnowledge and contextRuntime and executionEvidence and monitoring

Use cases

All use casesProduction incidentRelease preparationHotfix to productionSecurity scan triage
Why NopsAIIntegrationsSecurity

Resources

All resourcesAI agent governanceMCP governanceMCP securitySelf-hosted platforms
PricingGitHub

Company

How a run worksAboutContactBook a demo

What NopsAI is

NopsAI is a self-hosted automation control plane that executes YAML-defined workflows through Docker or Kubernetes runners.

ConceptNew userAdministratorDeveloper

Key points

  • A pipeline can combine deterministic shell scripts, LLM-backed goals, reusable steps, child pipelines, and human approval gates in one graph.
  • Runtime variables, encrypted secrets, governed Knowledge Context, Models, Agent roles, and MCP Profiles are all resolved before a step executes.
  • Runs start manually, from a schedule, from a GitHub App event, from a Git webhook source, from an external API trigger, or from a parent pipeline.
  • Final deliverables can be Markdown, JSON, HTML, PDF, Excel, or a dashboard publication, and are stored separately from raw task logs.
  • Everything is self-hosted: there is no NopsAI-operated cloud component in the current repository.

How it works

The product is GitOps-friendly by design. Pipelines, steps, schedules, triggers, scopes, knowledge, dashboards, and access can all live in a configuration repository, while durable execution, audit, credential, monitoring, and setup state stay in PostgreSQL.

LLM-backed work runs inside the per-run agent. There is no separate always-on LLM service; the agent calls the provider selected by the resolved Model.

The UI, CLI, REST API, hosted MCP surface, and runners all pass through the same authentication, authorization, and audit boundaries. No interface has a private path around AAA.

Limits

Current behavior
  • The repository defines no cloud-provider-specific infrastructure automation. Cloud installs treat NopsAI as a portable Kubernetes workload.

Implementation evidence

  • doc/architecture-overview.md

    Component map and deployment shape.

  • doc/feature-reference.md

    Functional capabilities exposed by the codebase and UI.