DRAGOPS
DRAGOPS
Documentation

Overview

An AI Ops platform. Build your own Harness — a workload-specific operational system you own, run, and govern in production.

DRAGOPS is an AI Ops platform. Your organization builds its own Harness — a workload-specific operational system composed from a visual node graph, integration packs, your own patterns, and a fault-tolerant runtime. You own it. You run it. You can take it with you.

You build patterns by wiring nodes together in a drag-and-drop editor. You author them yourself, or you bring your own AI agent (Claude Code, Claude Desktop, Cursor, ChatGPT) and have it author for you over Model Context Protocol (MCP). The result is the same: a real, visual, inspectable pattern that runs deterministically — not a prompt template, not a black box.

What is a Harness?

A Harness is the workload-specific operational system you build on DRAGOPS. It composes:

  • DRAGOPS Core — the platform: editor, primitives, runtime, connections, deployments
  • DRAGOPS [Use Case] bundles you load — vertical packs we ship: GRC, Detection Engineering, SOAR, etc.
  • Your patterns — the automation you build for your specific operations
  • Your connections, your data, your deployments — everything you wire up

A Harness is what you deploy. It is the unit of value. It is yours.

Key capabilities

  • Visual scripting — Build automations by wiring nodes together on a canvas. The graph is the source of truth, readable by anyone on your team.
  • Comprehensive node library — Hundreds of built-in primitives plus pack nodes for third-party services. HTTP, JSON parsing, string and array manipulation, flow control, math, date/time, and more.
  • BYO-agent authoring (MCP) — Use the AI agent you already pay for. Connect it to DRAGOPS via Model Context Protocol; have it compose, validate, and deploy patterns inside your Harness.
  • Multiple trigger types — Start patterns with webhooks, cron schedules, queue messages, log events, or calls from other patterns.
  • Pattern composition — Break complex operations into smaller patterns and call them like typed functions.
  • Real-time execution logs — Watch your pattern execute node by node, with timestamped values, branches, and errors.
  • Production-grade runtime — A fault-tolerant, supervised, distributed execution engine. Designed for sustained concurrent workloads — not just demos.
  • Self-hosted or SaaS — Same product, same containers. Run DRAGOPS in our cloud or on yours; your Harness runs identically.
  • REST API + OpenAPI — Every operation in the editor is available through the API.

Explore the documentation

On this page