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.

We’re always trying to improve our platform for our customers. Here, you can get an inside look at what’s currently in beta at Elementum. Remember, this list is provided as a heads-up—not as an invitation to test or a promise of what capabilities may be included. Most customers will not see these features until general availability. For what’s available today, see the most recent release notes.

​
What to expect

Heads-up, not a commitment. The information below explains how we treat beta features and why delivery and availability can change.
  • Currently being tested. These features are in active development and testing. Behavior and scope may change.
  • No guaranteed delivery date. We do not commit to a specific release date for any feature listed here.
  • No guarantee of delivery. Priorities shift based on resources and customer needs. A feature in beta may be delayed, changed, or not released to General Availability.
  • Limited availability. Some features may be tested by select customers before being released to General Availability. Broader access is not guaranteed until a feature is generally available.

​
Features in beta

This list is updated on a biweekly cadence to give you an idea of upcoming improvements likely to occur in Elementum. The colored badges indicate the area of Elementum where you will see the update.
​
Accept Files Through Run Agent Task

​
Accept Files Through Run Agent Task Intelligence

The Run Agent Task automation action can now pass files directly to an agent for processing. Use it in place of the AI File Reader to get broader model support and expanded file-handling functionality. Learn more about Run Agent Task.
​
Cross-App Skill Sharing

​
Cross-App Skill Sharing Intelligence

When configuring skills on an agent, select Custom access to pull in active skills from any App in your organization—not just the agent’s own. Agents discover and use cross-app skills the same way they use local ones, and deployed agents automatically include the Apps that provide their configured skills. Build a skill once and reuse it everywhere—agents can tap into capabilities across your entire org without duplicating configuration. Learn more about agent skills.
​
Dynamic Dropdowns on Agent-Driven Forms

​
Dynamic Dropdowns on Agent-Driven Forms Intelligence

Configure the Ask User Question tool with dynamic data sources so dropdowns populate from live platform data instead of static lists. Builders choose which field is rendered to users and whether the prompt accepts a single or multi-select response, and users can search within long dropdowns to find options quickly. Agents can now present real-time, data-driven choices to users—no more hardcoded option lists that go stale. Learn more about the Ask User Question tool.
Who can configure it: App Administrators.Where to find it: App → Agent → Configure → System Tools → Ask User Question.Configuration steps:
  1. When configuring an agent on an object, open the Ask User Question system tool.
  2. In the Dynamic dropdowns section of the popup, click Add Source.
  3. Choose the target object from the Records to Show dropdown.
  4. Choose the field that will be rendered to users in the dropdown.
  5. Choose whether the prompt accepts a single-select or multi-select response.
  6. To add multiple sources, click Add source in the top-right corner of the Dynamic dropdowns tile.
  7. Click Save when finished.
User experience:When a user is presented with a question that requires a dynamic dropdown input, they can use a search box to locate the correct answer. The selected value is then populated in the new record.
​
Voice Agent Gateway

​
Voice Agent Gateway Org Settings

Assign a single phone service to an Agent Gateway so one phone number becomes the entry point for multiple voice agents. Inbound calls flow through a pre-agent pipeline that authenticates the caller, detects language, and classifies intent to route to the right voice agent. Configure voice overrides on each agent to give them a distinct persona. When a caller’s intent shifts mid-call, the gateway detects it and transfers them to a different agent—no re-authentication required, with full transcript context carried over. The same multi-agent orchestration available in chat is now available on the phone. Learn more about Agent Gateways.
Who can configure it: Organization Administrators.Where to find it: Settings icon Org Settings → Agent Gateways (under Platform).Prerequisites:
  • An Agent Gateway with at least one agent added and routing rules defined.
  • A configured phone provider (for example, Elementum Twilio or a SIP trunk-enabled provider).
Configure the gateway:Follow the steps in Agent Gateways to create the gateway, add agents on the Agents tab, and define how requests are routed on the Rules tab. The Settings tab now includes a Voice section for connecting one or more phone numbers to the gateway.Connect a phone number:
  1. Open the gateway and select the Settings tab.
  2. In the Voice section, click + Add phone number.
  3. In the Create Phone Service dialog, configure:
    • Number Type — Select how the number is provisioned (for example, Create Number).
    • Area Code — Enter a U.S. toll-free area code (for example, 833, 866, 877, or 888).
    • Phone Provider — Select the telephony provider that will host the number (for example, Elementum Twilio).
    • Caller Access — Choose who can reach the gateway (for example, Allow All Callers).
    • Default Language — Set the primary language used for caller interactions.
    • Additional Supported Languages — Optionally add languages the gateway can detect and respond in mid-call.
    • Generate First Message — Toggle on to have the gateway automatically generate the opening message when a call connects.
  4. Click Save.
Connected phone numbers are listed under the Voice section. Add more than one phone number to give the gateway multiple inbound entry points.Behavior notes:
  • Inbound calls are authenticated, language-detected, and intent-classified by the gateway before being routed to the right voice agent based on the gateway’s rules.
  • When a caller’s intent shifts mid-call, the gateway transfers them to a different agent without re-authentication and carries the full transcript across the handoff.
  • Configure voice overrides on each individual agent to give them a distinct persona on calls routed through the gateway.
​
Linked Record Approvals

​
Linked Record Approvals Apps

Run an approval template multiple times on a single record, with each instance tied to a specific linked record. This lets teams handle sign-off per line item—such as one approval per purchase order, budget line, or deliverable—instead of bundling everything into a single record-level approval. Approval automation actions now also support a tag field, so automations can loop through linked records and target specific approval instances independently. Learn more about approvals.
Who can configure it: App Administrators.Where to find it: App → Approval Processes.Configuration steps:
  1. In your app, open Approval Processes and click Create New.
  2. Toggle on Linked Record Approval.
  3. Select the target linked object the approval should run against.
  4. Configure the remaining approval details (name, stage, approvers, and rules) as you would for a standard approval process.
Once enabled, the approval template can be run multiple times on the same record—once per linked record—either manually or through an automation.Manual requests:When Linked Record Approval is turned on, requesting approval on a record displays a Linked Line Item Record field. Choose the specific linked record from the dropdown to start an approval instance tied to that line item.Automated requests:Use the Start Approval Process automation action to trigger approvals based on business rules:
  1. Select the Record the approval applies to.
  2. Set Approval Process to the linked-record-enabled template (for example, Line Item Approvals).
  3. In the Linked [Object] Record field (for example, Linked PO Line Items Record), choose the linked record from the dropdown.
  4. Enter an Approval Process tag to uniquely identify this approval instance. Tags allow automations to differentiate between each approval chain on the same record—using the linked record’s ID is recommended.
Pair this action with a loop that iterates over linked records to start an independent approval instance for each one.
​
Organization Model Migration

​
Organization Model Migration Org Settings

Organization Administrators can select any AI model and migrate every usage to a different model in a single action. All automation actions and agents referencing the source model switch over automatically, making it easy to swap models org-wide without updating each agent and automation individually. Learn more about AI Services.
  1. On the Providers page, click the dropdown next to the service you would like to replace.
  2. Click the Replace Model icon Replace Model icon.
  3. Select the new model from the dropdown.
    In the popup, you can see all automations using the current model that will be affected by the change.
  4. Click Replace.
This may take some time depending on how many automations are using the previous model. Check progress in Background tasks if desired.
​
AI Provider Failover

​
AI Provider Failover Org Settings

Configure failover providers so traffic automatically reroutes if the primary is unreachable and the backup is active.
Who can configure it: Organization Administrators.Where to find it: Organization Settings → AI Services (under Platform) → Providers.Prerequisites:
  • At least two configured AI Providers of compatible model families.
  • Each provider must have active, tested credentials.
Configuration steps:
  1. Open the dropdown for the desired provider and click the Provider Details icon Provider Details icon.
  2. Click the Edit icon Edit icon next to Backup Providers.
  3. Select a provider to use in case of failover. Choose multiple to ensure several options are available.
  4. Select Save.
Behavior notes:
  • Failover is automatic — no user action is required during an outage.
  • Only providers whose status is Active are eligible targets.
  • Failover applies to all features consuming the primary provider (agents, search, summarization, etc.).
  • When the primary recovers, new requests resume routing to it.
​
SIP Trunking Provider

​
SIP Trunking Provider Org Settings

Bring your own telephony provider to Elementum by connecting SIP trunk-enabled phone providers to power voice agents.
Who can configure it: Organization Administrators.Where to find it: Organization Settings → Integrations → Phone.Prerequisites:
  • A SIP trunk-enabled account with a supported provider (for example, Twilio, Telnyx, or any standards-compliant SIP provider).
  • Credentials and the SIP URI/domain from your provider.
Configuration steps:
  1. Add a new phone provider and select SIP Trunk as the connection type.
  2. Enter a Name, add one or more SIP Gateways (the IP addresses or hostnames that will handle call routing), and select an Authentication Type.
    • IP Based — authenticate using the gateway’s IP address; no additional credentials required.
    • Credentials — enter a username and password.
  3. Use the provided callback URL to complete the provider configuration.
  4. Test the connection to confirm Elementum can register with the trunk.
  5. Assign inbound/outbound phone numbers from your provider to specific voice agents.
Behavior notes:
  • Using your own SIP trunk lets you keep existing carrier contracts and phone numbers while routing calls through Elementum’s voice agent stack.
  • Per-call audio, transcripts, and telemetry still flow through Elementum for logging and agent memory.
  • Outbound caller ID and number porting are handled by your telephony provider, not by Elementum.