Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.elementum.io/llms.txt

Use this file to discover all available pages before exploring further.

Agents are conversational AI components that operate within your apps. Unlike one-shot AI actions in automations, agents maintain context across a conversation, use tools to read and write data, and follow policies you define. A first-line support assistant is a typical starting point, but the same patterns apply to onboarding, sales qualification, request intake, or any process where a guided conversation speeds up routine work.

​
What agents can do

  • Tools and data access — Agents can query Elements, update records, trigger automations, and call external APIs.
  • Skills — Modular, reusable capabilities that agents discover and execute at runtime.
  • Starting actions — Pre-configured action chips that guide users toward common tasks without typing.
  • Multi-channel deployment — Run agents in your Elementum app, Microsoft Teams, Slack, or over the phone via Twilio.
  • Multi-agent coordination — Agents communicate using the Agent-to-Agent (A2A) protocol to hand off tasks and share context across workflows.

​
How agents fit with the rest of AI in Elementum

Agents sit on top of the same provider and AI service layer as the rest of the platform. Before you build an agent, make sure you have:
  1. At least one configured AI provider.
  2. An LLM service the agent can use as its model.
  3. (Optional) An embedding service if you want the agent to perform AI Search over your records.
Once those prerequisites are in place, the rest of the Build agents group walks through creating, configuring, and deploying an agent end-to-end.

​
Where to go next

Building Agents

Create, configure, and test agents in your apps

Agent tools and deployment

Tool types, deployment options, and external agents via App Intelligence

Agent Skills

Define reusable skill modules that agents discover and execute at runtime

Agent Starting Actions

Add clickable prompt chips that guide users toward common tasks

Agent Gateways

Create a unified org-wide entry point that routes requests to the right agent

Architecture and A2A protocol

Internal component model, A2A discovery, authentication, streaming, and multi-turn conversations

​
Deploy agents to channels

Agents can operate beyond the Elementum interface. Connect them to the communication channels your teams already use, or trigger agent conversations directly from automation workflows.

Microsoft Teams

Run agents as bots in Teams conversations

Slack

Deploy agents in Slack channels for team collaboration

Phone (Twilio)

Enable voice conversations with agents over the phone

Agent Tasks in Automations

Trigger agent interactions from automation workflows

​
External agents

In addition to building agents inside Elementum, you can connect specialized agents that run on external platforms. These external agents are configured at the App level through Intelligence settings and invoked from automations via the AI Provider you connected.

Snowflake Cortex Agents

Use Cortex Agents that run on your Snowflake data warehouse

AWS Bedrock Agents

Invoke Bedrock Agents hosted in your own AWS account