⌥ § Documentation · Field Manual

Documentation.

Full control, end to end — self-hosted, your clouds, your models, your decisions. New here? Start with the Quickstart, read why we built it, or browse the full subsystem reference.

Recommended provider

Local-first, built on Ollama.

OpenAgentics stands up end to end on a local Ollama — chat and embeddings — with no cloud account and no API keys. It's the provider we develop and test against every day.

Default provider Bootstrap seeds Ollama on first boot — your stack chats with zero credentials.
Developed & tested against it Over 200 test files reference Ollama across unit, integration, and end-to-end suites.
Embeddings by default nomic-embed-text ships in the compose and powers the semantic tool router out of the box.
Free, offline, no keys Open-weight models run on your own hardware. Air-gap it — nothing leaves your box.

Read the Ollama story ollama.com Model library GitHub

Facts about the OpenAgentic source — not an official partnership or endorsement by Ollama. Ollama is our recommended provider.

Quickstart

Install OpenAgentics and have it running locally in a few commands.

API Reference

The Fastify + OpenAPI 3.1 HTTP API — chat, MCP tools, agents and models, rendered live.

Subsystem Reference

Every subsystem — MCP servers, flows, agents, security, infra — generated from the source.

⌂

Runs on your infrastructure

Self-hosted or air-gapped — not hidden behind a SaaS, not locked to any vendor.

⊙

Your IAM, your credentials

Cloud tools authenticate through scoped roles you grant; tools run as you, never as root.

✋

Human-in-the-loop

Nothing changes your systems without approval — every action checksummed into an audit log.

Getting started

Quickstart →

Get OpenAgentic running — Docker Compose on one host, or Helm on a Kubernetes cluster.

Architecture →

How OpenAgentics fits together — the real services, data plane, and the systems inside the agent core.

Concepts

Chat →

Control your whole stack from chat — run workflows and agents, read metrics and logs, and stand cloud resources up or tear them down on any criteria, behind a human-approval gate on writes.

Flows →

Visual node-based workflow builder with branching, retries, signed traces, and governance.

Providers & models →

Local-first on Ollama — free, offline, no API keys — with every major cloud provider also supported through one canonical streaming format.

Work with your real repositories through the built-in GitHub tool, with your own credentials — the agent reasons over code and runs it in a sandbox, behind the same approval gate and audit trail as any other action.

Deploy

Docker Compose →

Single-node deployment and local dev with Docker Compose.

Kubernetes →

Deploy OpenAgentics into any Kubernetes cluster with the self-contained Helm chart.

Helm + Vertex AI →

Deploy OpenAgentic on Kubernetes with the in-tree Helm chart, backed by Google Vertex AI (Gemini).

Reference

MCP servers →

The MCP servers OpenAgentics ships, their authentication models, and tool-level policy.

Configuration →

How OpenAgentic is configured — model provider, .env / Helm bootstrap, and the setup wizard.

Troubleshooting →

The eight failure modes you're most likely to hit on a first deploy — symptom, cause, fix.

Enterprise

Identity & OBO (enterprise) →

Per-user identity — running each action as the signed-in human via SSO and on-behalf-of token exchange — is an Gnomus enterprise feature. The open-source edition is local-auth only. This page documents the enterprise identity architecture.

Fleet & Mission →

Where Gnomus is headed for enterprise AIOps — the Mission and Fleet orchestration layer above today's production-ready workflows.