Autonomous Database Operations Within Defined Boundaries
Create named AI agents with their own API keys and pre-approved query access. Every action is logged in an immutable ledger. Agents execute pre-approved query patterns, take actions based on data triggers, and manage routine data operations without human intervention.
AI Agents
each with its own identity
Key Management
per-agent credentials
Query Patterns
approved before execution
Immutable Audit
every action logged
How it works
Set up your ontology, create pre-approved queries, then give each agent exactly the access it needs.
Configure Ontology Forge
Set up Ontology Forge to extract your business semantic layer. This gives agents the context they need to understand your data.
Create Pre-Approved Queries
Define the query patterns that agents are allowed to execute. Each query is reviewed and approved before any agent can use it.
Add Named Agents
Create agents with their own API keys and assign which pre-approved queries each one can access. Distribute keys and connect via MCP.
What you get
Named Agent Management
Create multiple named AI agents, each with its own identity and scoped access. Agents authenticate as themselves, not as a generic connection.
Per-Agent API Keys
Each agent gets its own API key. Rotate, revoke, or monitor individual agent credentials without disrupting others.
Pre-Approved Query Patterns
Define exactly which queries each agent is allowed to execute. No ad-hoc access — every operation is pre-approved.
Immutable Action Ledger
Every agent action is logged with full context: who, what, when, and which pre-approved query was used. Complete audit trail for compliance.
Use it when
You need AI agents running routine data operations 24/7 without human bottlenecks.
You want every agent action logged in an immutable ledger for full auditability.
You need agents to operate within defined boundaries, executing only pre-approved query patterns.
Multiple AI agents need their own identities and scoped credentials rather than sharing a single connection.
Why this matters: Without pre-approved query boundaries and action ledgers, autonomous agents can execute unintended operations at machine speed. The blast radius of a misconfigured agent is orders of magnitude larger than a human mistake. AI Agents & Automation ensures every agent operates within defined guardrails.
Start with named agents today
Set up your ontology and pre-approved queries, then create named agents with scoped access. The AI Agents app is available now in Ekaya Central.