How to Use the ORIA Reasoner

The ORIA Reasoner is one of the tools available from My Tools on the ARPIA Home screen. It lets you explore your organization's knowledge nodes with Oria — short for Ontology Reasoning Interactive Agent — an agentic assistant that can answer questions, run semantic and ontology queries, draw on your organization's institutional memory, and build visualizations and applications directly from your data.

ORIA Reasoner is designed as a full tool for end business users, not just technical or admin staff — no query language or ontology expertise required to ask a question and get a grounded answer.

Overview

Use Cases

ORIA Reasoner is built for exploring relationships in your data, not just retrieving records — so it's most valuable in situations where the connections between things matter as much as the things themselves.

• Knowledge discovery for new team members. Instead of asking a colleague "how does X relate to Y" or hunting through scattered docs, a new hire can ask Oria directly (e.g., "what data nodes feed into the executive dashboard?") and get an answer grounded in the actual ontology.
• Governance and compliance mapping. Before changing an access policy or data flow, ask Oria to trace dependencies (e.g., "what systems or reports depend on this node?") to understand blast radius before making the change.
• Cross-domain analysis. When a question spans multiple areas (CX, security, operations, finance), Oria can query across the ontology nodes you have access to and surface connections a single-domain search wouldn't reveal.
• Pre-change impact checks. Before deprecating a doc, table, or workflow, ask what's connected to it — this surfaces hidden dependents before you break something downstream.
• Answers grounded in prior decisions. Because Institutional Memory is on by default, Oria answers from your organization's recorded decisions, rules, and procedures — not only from what's in the data or from general knowledge.
• Building visual artifacts for stakeholders. Use a session to generate a dashboard, relationship analysis, or timeline you can pin as an Artifact and share in a meeting or report.
• Ad-hoc investigation. During an incident or audit, spin up a session scoped to just the relevant nodes and ask targeted questions instead of manually piecing together context from multiple tools.
• Operational tooling without a development cycle. Use Live Apps and Repository Storage to stand up a working lookup, reconciliation view, or query builder over your session data without waiting on a build request.

Before You Start

ORIA Reasoner requires a one-time setup by an ARPIA administrator at the work-area level. You don't configure this yourself.

If the tool is greyed out, or the model picker inside a session is disabled, a Model Group hasn't been set for the Reasoner. Ask your ARPIA administrator to configure it.

Three things you may find switched off or empty are also administrator-controlled, and are not errors:

Administrators: see Configuring the ORIA Reasoner for the full setup procedure.

Getting Started

1. From ARPIA Home, select ORIA Reasoner — either under My Tools, or from the Side Menu on the left.
2. You'll land on the Oria workspace. With no sessions yet, the left panel shows No sessions yet, and the center reads:

   Explore your data with Oria — Open a session from the left, or start a new one. You can also just type below — your first question creates the session for you.

Two ways to start

Tip: If you start by asking, rename the session afterward to reflect the question. Sessions are listed for later reference, and a clear name saves time when you come back to it.

The new-session state

Once a session exists, the center panel guides you through setup with three Next steps cards:

None of these is mandatory before asking. If you ask without configuring anything, Oria will tell you what it needs.

The Session Workspace

The workspace is organized into three areas.

Left Panel — Sessions

• Search sessions to find previous work. Sessions are grouped under Recent Sessions with a relative timestamp and an open link on each card.
• Scheduled Tasks: you can save scheduled query definitions here. As of this writing, the scheduling engine is not yet enabled — definitions are saved and will run automatically once the engine is activated.

Center Panel — Oria Chat

• Type your question in the "Ask about your data…" box.
• You can attach files or switch the underlying Model used for the session from the controls at the bottom of the chat box.
• Answers are shown with their supporting evidence. When Oria draws on Institutional Memory, the response includes a Memories Used block listing each memory ID, its title, its Relevance (primary or supporting), why it was used, and any external references it cites. Read this block as-is — it is how you verify what the answer was actually grounded in.

Right Panel — Session Configuration

This panel controls what Oria has access to and produces within the session. It is visible even before a session is open, though its controls only apply once one exists.

Institutional Memory has its own guide. See How to Use Institutional Memory (Cerebro) in the ORIA Reasoner for scoping, saving, and verification practice.

Institutional Memory

Institutional Memory (Cerebro) lets Oria answer from your organization's recorded decisions, rules, and procedures — not just from the data in your session or from the model's general knowledge.

It is enabled by default in every session, scoped to every domain your profile can access. You do not need to turn it on. What you may want to do is narrow it, or turn it off.

Two controls sit in the panel header:

The Institutional memory dialog

Institutional Memory is a read and write path. Oria can record new memories during a session, not only consult existing ones. Shared-domain writes always prompt for confirmation; personal-domain writes do not, unless you enable Ask me before saving anything.

This page covers the panel only. For domain scoping strategy, how to phrase questions so memory answers well, how to save from a session, verification practice, and known limitations, see the dedicated guide:

📘

How to Use Institutional Memory (Cerebro) in the ORIA Reasoner

Read it before you rely on a memory-grounded answer in a decision, a deck, or a committee record.

Adding Ontology Nodes

Click + next to Ontology Nodes to open the picker.

Three ways to find what you need:

• Search nodes by name — free-text search, with a filter control alongside it.
• AI recommender — describe what you want to solve in the "Or describe what you want to solve" box and click Recommend; Oria proposes the best nodes to load for that problem.
• Browse — scroll the available node cards.

Each node card shows its name and a description of what the node contains, what one row represents, and which fields are authoritative. Read the description before selecting — it tells you the grain of the data (for example, one row per order line rather than per order) and any caveats about specific columns. Getting this wrong is the most common source of misleading answers.

A running n selected count appears above the cards. Click Save to attach the selected nodes to the session, or Cancel to discard.

Node descriptions are the contract between you and the answer. If a description is missing or vague for a node you rely on, raise it with the node's owner — Oria reasons from that description as much as from the data itself.

Session Tools

Click the + next to Tools in this session to open the Session Tools dialog.

The agent can only see tools you enable here. Nothing in this dialog is on by default. If Oria says it can't do something, check this dialog first — the capability is usually present but not enabled for the session.

Two ways to find what you need:

• Search — the "Search tools, actions or nodes…" box filters across every tab at once.
• Recommend — describe what you're trying to do in plain language and click Recommend; Oria proposes the tools, actions, and MCP connections to enable for that task.

Tools are organized across five tabs.

Internet Tools

Artifact Tools

Artifact Tools are what Oria uses to build something — a visualization, an application, or a working environment — rather than just answer in prose.

Repository Storage is a two-step setup. Adding the tool is not enough — after adding it, open its ⚙ icon in Tools to create the database and attach it. Until you do, the tool is present but has nothing behind it.

Write-back is governed. Work Boards are read-only unless a governed Ontology Action exists to write the status back, and Oria Machine runs with human-in-the-loop approvals. Enabling a build tool does not by itself grant write access to your ontology.

Ontology Actions

Governed actions your profile is permitted to invoke against ontology nodes — this is the mechanism by which a session can write back rather than only read.

If you see "No actions granted to your profile," no actions have been granted to you. Contact your ARPIA administrator.

MCP Collections

Internal ARPIA MCP / API collections that let the agent query integrated systems.

If you see "No MCP Collections are granted to your profile," ask your ARPIA administrator to grant you access to a collection.

Connected Apps

External MCP connectors — third-party systems connected to your work-area.

If you see "No external MCP connectors are available to your profile," ask your ARPIA administrator to add a connector and grant your profile access.

Click Save to apply your selections, or Cancel to discard them. Tool selections are session-specific — enabling a tool in one session does not enable it in another.

Typical Workflow

1. Start a session — either by clicking +, or simply by asking your first question.
2. Add the relevant Ontology Nodes, reading each node description to confirm the grain of the data.
3. Narrow Institutional Memory to the domains that bear on the question, if the default (all domains) is too broad.
4. Open Tools in this session and enable only what the question needs — or describe the task and click Recommend.
5. (Optional) Upload supporting Files.
6. Ask your question.
7. Check the Memories Used block and source references before acting on the answer.
8. Pin useful results to Artifacts for later reuse.

Scope narrowly. Adding every node, every domain, and every tool to a session degrades answer quality rather than improving it. Start with what the question actually requires, and widen only if the answer comes back incomplete.

Known Limitations

• Requires admin setup: ORIA Reasoner is disabled until an admin configures a Model Group for it. This is set once at the work-area level, not per user — see Before You Start.
• Tools are opt-in per session: no tool is enabled by default, and tool selections do not carry over between sessions.
• Institutional Memory scope is opt-out, not opt-in: sessions default to every domain your profile can access. Narrow it deliberately when scope matters.
• Personal-domain memory writes are silent by default: enable Ask me before saving anything if you want to review everything Oria commits to your personal memory.
• Three tool categories are permission-gated: Ontology Actions, MCP Collections, and Connected Apps will appear empty until an administrator grants your profile access. This is expected behavior, not an error.
• Repository Storage requires a second step: the database must be created and attached from the tool's ⚙ icon after the tool is added.
• Scheduled Tasks: the scheduling engine is not yet active. Scheduled definitions are saved but will only begin running once the engine is turned on.