What is Nenjo
An introduction to Nenjo — what it is, what makes it different, and how its building blocks fit together.
Last updated: June 15, 2026
Nenjo is a zero-trust platform for building and running teams of AI agents — together, on your terms.
Most AI tools ask you to hand over your data and trust them with it. Nenjo is built the other way around. Your prompts, documents, chats, and agent responses are encrypted on your own devices before they ever leave them, so the platform can store and route your work without ever seeing it. It is collaborative, so your team builds and operates these agents together. And it is modular, so the same small set of building blocks can power anything from a personal assistant to a production-grade AI application.
Nenjo runs on a hybrid architecture: a hosted dashboard and backend for convenience, and an open-source agent harness you run on your own computer or infrastructure for flexibility and control.
What makes Nenjo different
Zero-trust
The platform never sees your plaintext. All chat messages and dashboard-authored content — agent prompts, knowledge documents, agent responses — are encrypted client-side. Material that may contain sensitive business or personal information is unreadable to Nenjo by design.
Your devices and your own Nenjo harnesses are the trust boundaries: they are the only places your content exists in the clear. Everything in between simply stores and routes encrypted data for your convenience. For how this works, see Security.
Modular
Nenjo was built from the ground up to be composable. Instead of one monolithic assistant, it gives you a small set of primitives — projects, agents, abilities, domains, routines, context blocks, and knowledge — that build on each other.
These pieces combine into workflows: repeatable processes that route work through the right agents, tools, and approval steps. The same building blocks let you assemble many different kinds of workflows, from a single-step assistant to a multi-agent production pipeline. And you do not have to wire them up alone — Nenji can help you design and build these workflows for you.
You can start with a single project and one agent, then layer in structure as the work gets more serious. The same building blocks scale from a quick personal helper to a production AI workflow, so you never outgrow the platform or have to start over.
Collaborative
Nenjo is built for teams. Create a workspace, invite your teammates, and design, operate, and control your agentic workflows and applications together — with shared knowledge, clear roles, and explicit permission boundaries instead of scattered chat histories and one-off prompts.
Hybrid
Nenjo splits the work between a hosted layer and a layer you own. The dashboard and backend are hosted, so you get the convenience of a managed product. The agent harness itself is open source and self-hosted, running on your own computer or infrastructure.
This is a deliberate balance of flexibility, convenience, security, and extensibility: you get the polish of a hosted platform without giving up control of where your agents actually run.
Meet Nenji
At the center of the platform is Nenji, the default guide and orchestrator. Nenji helps you create resources, explains how they fit together, inspects existing configuration, and moves between high-level design and concrete platform actions.
When you know the outcome you want but not yet which resources should exist, Nenji helps you shape the workspace — asking clarifying questions, drafting configurations, and connecting the pieces. See Nenji for how to work with it.
The core model
Each building block has one job, and together they compose into the workflows above. A project is a container for a unit of work: it scopes the runtime, memory, and context an agent operates within, so every effort runs with exactly the state it needs and nothing else. Agents do the work. Routines turn that work into repeatable workflows. Abilities give agents focused specialist tools, and domains let users intentionally switch an agent into a different mode when the work needs different guidance, permissions, or auditability. Durable source material lives in the Library, separate from any single project, so it can be shared and reused across your workspace.
You can start with only a project and an agent. As the work becomes more serious, add structure where it helps: routines for repeatability, abilities for specialist work, domains for explicit high-trust modes, and Library knowledge for durable source material.
For the resource-by-resource view, see Building Blocks.
Where to start
If you are new to Nenjo, start with the building blocks, then create a small project and one agent. Add structure only when the work asks for it.