Estimated read time: 6 minutes

Overview

AI agents are intelligent connectors that link AskElephant to the tools your team already uses—like your CRM, project management software, or communication platforms. Instead of manually pulling information from multiple places, agents retrieve data automatically and let AI work with it in real-time.

Think of agents as bridges between AskElephant and your business. They pull information from one side, bring it into AskElephant where AI can analyze it, and then push results back out. This means you have all the context you need without switching between apps.

Key Terms

Agent: An AI-powered connector that retrieves, reads, updates, or creates information in external tools and platforms.

Integration: A connection between AskElephant and another business tool (like HubSpot, Salesforce, or Slack).

External Data: Information stored outside of AskElephant—in your CRM, project management tool, or other platforms.

Real-Time Access: The ability to pull current information from connected tools as your workflow runs.

Context: Background information about a customer, deal, or project that helps AI provide better analysis.

What Agents Do

Agents perform four core actions:

Read: Pull information from connected tools. For example, an agent might retrieve a customer's deal stage from Salesforce or pull a task list from Asana.

Create: Add new information to external tools. An agent could create a new opportunity in HubSpot or add a task to Monday based on meeting insights.

Update: Modify existing information. If a meeting reveals a deal has moved to a new stage, an agent can update that in your CRM automatically.

Delete: Remove information (some integrations restrict this functionality.)

Because agents can read and write data, they give AI access to the full picture of what's happening in your business.

Available Integrations

Agents connect AskElephant to the platforms your team uses most:

• HubSpot: Pull and update contacts, deals, companies, and custom properties
• Salesforce: Access opportunities, accounts, and related records
• Monday: Read and create tasks, update project status
• Asana: Manage projects, tasks, and timelines
• Linear: Track issues and development work
• Notion: Read databases and pages, create new entries
• Slack: Send notifications and messages to channels
• Google Calendar: View upcoming events and meeting details
• Microsoft Calendar: Access calendar information from your organization

If a tool isn't listed, reach out to the Herd—they may have built workarounds or have integration requests underway.

How Agents Work in Workflows

Agents are most powerful inside workflows. Here's a typical flow:

Step 1: Your workflow triggers on a new meeting or other event.

Step 2: An agent node runs and pulls relevant data from a connected tool. For example, an agent searches HubSpot and finds all deals associated with the meeting attendees.

Step 3: AI analyzes the combined data. Now AI has both the meeting transcript and the customer's deal history.

Step 4: Results are created or updated. The agent can then push insights back—updating deal notes, creating a task, or sending a Slack message.

All of this happens automatically without manual data entry.

Common Agent Use Cases

Sales Prep

Before a customer call, an agent pulls deal information from your CRM and past meeting notes. AI synthesizes this into a pre-call summary showing what's been discussed, where the deal stands, and what risks or opportunities exist.

Sales to Customer Success Handoff

When a deal closes, an agent retrieves the entire sales engagement history from your CRM. AI creates a comprehensive handoff document highlighting customer pain points, feature excitement levels, and recommended onboarding focus areas—then updates the CS team's project management tool automatically.

Customer Health Monitoring

An agent pulls meeting transcripts, support tickets, and deal information. AI flags churn risks, expansion opportunities, or escalations that need immediate attention. Results can be pushed to Slack for real-time team alerts.

Action Item Tracking

After a meeting, an agent creates tasks in Asana or Monday based on discussion points AI identified. These are linked to the relevant customer record in your CRM, keeping everything connected.

Account Intelligence

Before a prospect call, an agent searches your CRM for any past interactions with the company or similar prospects. AI uses this to inform conversation strategy and flag what competitors or solutions have been discussed.

Agents vs. Tools vs. Workflows: When to Use Each

Quick guide:

• Need to pull info from HubSpot? Use an agent.
• Want to search past meetings quickly? Use a tool.
• Building a complete automated process? Use a workflow with agents and tools combined.

Using Agents in AskElephant Chat

You don't need a workflow to use agents. They're available directly in conversations.

In a meeting conversation, you can:

• Ask an agent to pull customer information: "Show me this contact's deals in HubSpot"
• Request updates: "Create a task in Asana for follow-up"
• Combine agent data with AI analysis: "Based on this meeting and their account history in Salesforce, what should we prioritize?"

Agents in chat are helpful when you need information on-demand without setting up full automation.

Getting Started With Agents

Step 1: Connect Your Tools

In your workspace settings, authenticate the integrations you want to use. This typically requires logging in to each platform once.

Step 2: Use Agents in Conversations

Start asking for information from connected tools in meeting chats. Explore what data is available.

Step 3: Build Workflows With Agents

Once you're comfortable, incorporate agents into workflows for full automation. Start simple—pull one piece of data, run an analysis, and send results somewhere.

Step 4: Iterate

Test workflows on a small scale first. Once you confirm they work, expand to your full team.

Important Notes

Permissions Matter: Agents can only access data your AskElephant account has permission to see in the connected tool. If you don't have access to a deal in Salesforce, an agent won't retrieve it.

Data Privacy: All data pulled by agents stays within AskElephant unless you explicitly configure it to be shared elsewhere. Review your integrations to understand what's being accessed.

Troubleshooting Common Issues

Agent can't find the data I'm looking for

Make sure the data exists in the connected tool and that you have access to it there. Check that you're searching with the right parameters—exact names or IDs often work better than partial matches.

Agent returned incomplete information

Some integrations have limits on what data they can access. Review the integration documentation or reach out to the Herd for workarounds.

Integration stopped working

Connection credentials can expire. Reconnect the integration in your workspace settings to refresh the authentication.

Next Steps

Start by exploring agents in your AskElephant chat. Pull information from one connected platform and see what becomes available. Once you're comfortable with what data is accessible, consider building a simple workflow that automates a repetitive task.

Questions about specific integrations? Connect with the Herd in our community forum. They've connected every platform and can share what works best for your use case.

Need Additional Help?

If you have questions or need further assistance, the AskElephant support team is here to help!

You can reach our support team in several ways:

• click the chat button in the bottom right corner of your screen,
• email us at support@askelephant.ai
• or use @askelephant support in your dedicated Slack channel.

We're committed to getting you the answers you need as quickly as possible.