Data Explorer
Learn about what the Data Explorer Agent does and when to use it.
The Data Explorer agent is a read-only assistant that helps you interact with your data through natural language queries. It retrieves entity details, matches, relationships, and tenant metadata without modifying them.
Who is it for?
Data Steward
Business User
Reltio Configurator
Solution Architect
For more information, see About roles.
Why would I use it?
Use the Data Explorer agent to investigate records, review entity connections, examine match activity, and gain insights about your tenant's structure and lineage without risking data changes.
When and where would I use it?
Use this agent when you need to:
- Explore entity profiles or relationship graphs
- Investigate potential duplicates and match statistics
- Understand your tenant’s data model, configuration, and permissions
The agent runs in AgentFlow through the MCP Server in read-only mode.
How does it support business goals?
It increases user confidence and transparency by making data accessible without compromising integrity or requiring advanced query languages. It enables governance teams to explore metadata safely.
Core capabilities
- Entity search and retrieval
- Relationship graph navigation
- Match score analysis and duplicate tracking
- Tenant data model inspection
- Merge history and lineage reporting
- Interaction and timeline exploration
- Audit trail and permission visibility
Inputs and outputs
| Inputs | Outputs |
|---|---|
| Natural language query: Entity ID – Provide a unique identifier for a specific entity. For example, Attribute Filter – Use match conditions like equals, exists, missing, or range to search based on attribute values. Free Text Query – Enter natural language prompts such as “Find John Smith in California.” Pagination Settings – Define how many results to return, where to start, and how to sort them. Field Selection – Specify which attributes to include in the response, such as name, email, or all available fields. Relation Parameters – Provide relation types, graph depth, or entity roles to explore relationships between entities. Match Filters – Set parameters like score range, match rule, or confidence level to refine match results. Model Request – Ask for data model definitions such as entity types, relation types, and survivorship rules. Audit Filters – Query the history of merges or unmerges using timestamps, event types, or user information. Interaction Filters – Filter interactions by type, status, or timestamp to retrieve activity timelines. Tenant Metadata Request – Request tenant-wide metadata, including data model structure and security settings. |
Entity Details – Full entity profile including URI, labels, attributes, and source system crosswalks. Search Results – List of matching entities with support for pagination, offset tracking, and result counts. Relationship Data – Connected entities with direction, relation type, and relationship attributes. Match Analysis – Summary of potential matches showing scores, confidence levels, and match rules triggered. Match Statistics – Tenant-wide statistics grouped by entity type, match rule, or confidence category. Data Model Details – Configuration for entity types, attributes, survivorship rules, and match logic. Merge History – Logged actions with timestamps, merged entities, users, and applied rules. Interaction Records – Timeline of interactions linked to the entity, filtered by type or date. Tenant Overview – Overview of tenant settings including model structure, roles, and permission definitions. |
Safeguards, permissions, and governance
- Read-only by design — cannot modify, delete, or create records
- Respects user roles and permission scopes
- Audits all queries and viewed entities
Limitations and edge cases
- Cannot perform updates, merges, or deletions
- Response scope is limited to user's data access permissions
- Processes one query at a time