Skip to main content

AI Agent Action: Access Full Conversation History

AI Agents can now read and use a contact’s full conversation history across all channels before responding.

Who This Is For / When to Use

Use this feature when you want your AI Agent to:

  • Respond with full context from past interactions

  • Avoid repeating questions already answered

  • Continue conversations across multiple channels seamlessly

  • Deliver more human-like, informed replies

What’s New

The AI Agent action in Workflows now has access to a contact’s complete conversation history.

This includes conversations across:

  • SMS

  • Email

  • WhatsApp

  • Instagram

  • Facebook Messenger

  • TikTok

  • Web chat and other connected channels

The AI Agent reads prior interactions before generating a response, ensuring continuity and accuracy.

What Data the AI Agent Can Access

For each conversation, the AI Agent can retrieve:

  • Channel where the conversation occurred

  • Message content (last 20 messages per thread by default, up to 50)

  • Message direction (inbound vs outbound)

  • Timestamps for each message

  • AI-generated summary of the full conversation

This allows the agent to quickly understand long or multi-channel histories.

How It Works

The AI Agent automatically accesses conversation history when needed to complete its task.

  1. A workflow triggers the AI Agent action

  2. The agent retrieves past conversations for the contact

  3. The system compiles:

    • Recent messages

    • Conversation threads across channels

    • A summarized context (for long histories)

  4. The AI Agent generates a response using this full context

Where This Appears in Workflows

The conversation history is accessed during AI Agent execution inside a workflow.

  • The AI Agent sits after triggers like:

    • Form submitted

    • Appointment status

  • It can then trigger downstream actions such as:

    • Send email

    • Send SMS

    • Create opportunity

    • Update contact

Execution Logs and Visibility

You can verify conversation history usage in workflow execution logs.

Look for:

  • “Get Past Conversations” step

  • Successful execution status

  • Timing and token usage

  • Final AI response output

Why This Matters

Previously:

  • AI Agents only saw messages from the current interaction

  • Context from other channels or earlier conversations was not available

Now:

  • AI Agents behave like a human team member reviewing history before replying

  • Conversations remain consistent across channels and time

Example:

  • A contact asks about a quote via email

  • Follows up later on Instagram

  • The AI Agent responds with awareness of both interactions

This improves:

  • Response accuracy

  • Customer experience

  • Conversion rates

Common Issues and Fixes

AI Agent is not referencing past conversations

  • Ensure the AI Agent action is included in the workflow

  • Confirm the workflow is triggered by a contact with existing conversation history

Limited message history retrieved

  • Default is last 20 messages per thread

  • System may extend up to 50 depending on context needs

No conversation data found

  • The contact may not have prior conversations logged

  • Verify channels are connected and syncing properly

FAQ

Does the AI Agent read conversations from all channels automatically?

Yes. The AI Agent pulls conversation history across all connected channels, including SMS, email, social messaging platforms, and chat.

How many past messages does the AI Agent use?

By default, the agent retrieves the last 20 messages per conversation thread, with the ability to expand up to 50 messages when needed.

Does the AI Agent summarize long conversations?

Yes. An AI-generated summary is included to help the agent quickly understand long or complex conversation histories.

Do I need to manually enable this feature?

No. This functionality is automatically available within the AI Agent action when used in workflows.

Can the AI Agent respond differently based on past interactions?

Yes. The agent uses conversation history to tailor responses, reference previous discussions, and maintain continuity.

Did this answer your question?