Skip to content
AI Employee

Conversation AI - Agent Logs

Learn how Agent Logs helps teams review Conversation AI activity, troubleshoot AI responses, inspect tool use, and monitor performance across contacts.

10 min read2,142 words19 explained images
View as markdownOpen in ClaudeOpen in ChatGPT
Each image has a one line explanation. Switch to full detail for the step it belongs to, the fields and buttons involved, examples and the whole procedure.
On this page
  1. Key Benefits of Agent Logs for Conversation AI
  2. Important concepts
  3. Conversation
  4. Turn
  5. Step
  6. Where to Access Conversation AI Agent Logs
  7. 1. Agent Logs Tab Under AI Agents (Universal View)
  8. 2. Conversations Page
  9. 3. Contact Records and Contact Side Panels
  10. Sessions View in Agent Logs
  11. Filtering Sessions
  12. Contacts View in Agent Logs
  13. Metrics in Agent Logs
  14. Customizing Metrics Layouts
  15. Log Detail View
  16. Raw Conversation and Execution Timeline
  17. Raw conversation
  18. Execution timeline
  19. Reviewing a Specific AI Message
  20. Tool and Action Calls
  21. Parsed and Raw Views
  22. Parsed view
  23. Raw view
  24. Latency and Response Timing
  25. Frequently Asked Questions
  26. Related Articles

Video 1 Open in a new tab

Agent Logs for Conversation AI helps teams review how AI handled a customer conversation, including the messages, actions, timing, and session details behind each interaction. This article explains where to access Conversation AI Agent Logs, what information is available, and how teams can use logs to troubleshoot responses, review contact activity, and monitor performance.

TABLE OF CONTENTS

What is Agent Logs for Conversation AI?

Agent Logs for Conversation AI gives users visibility into what happened during a Conversation AI interaction. Instead of only seeing the final message sent to a contact, users can review the conversation, the AI response, the actions the AI selected, and the steps it followed before responding.

Think of Agent Logs as a transparency view for Conversation AI. It helps users understand what their AI agent did, how it handled a customer message, and where each response came from.

Key Benefits of Agent Logs for Conversation AI

Conversation AI can perform different actions depending on the customer message, the configured agent instructions, and the tools available to the agent. For example, it may answer a question, check calendar availability, collect contact details, use a knowledge base, end a conversation, or hand the conversation over to a team member.

Agent Logs help users see these actions more clearly. They are useful when a user wants to understand questions such as:

  • What did the contact say?

  • How did Conversation AI respond?

  • Did the AI call the expected action or tool?

  • What information was passed into an action?

  • What did the action return?

  • Why did the AI take longer to respond?

  • Did the AI use the knowledge base?

  • Did the AI hand over or end the conversation?

Agent Logs are especially helpful because AI agents are not always linear. The same agent may take different paths depending on what the contact says. Logs make those paths easier to review.

Key Benefits of Agent Logs for Conversation AI (image 1 of 19) What this shows Agent Logs are especially helpful because AI agents are not always linear. What this shows Illustrates the "Key Benefits of Agent Logs for Conversation AI" section of "Conversation AI - Agent Logs". This screenshot appears in the "Key Benefits of Agent Logs for Conversation AI" section of "Conversation AI - Agent Logs". The text alongside this image reads: Agent Logs are especially helpful because AI agents are not always linear. The same agent may take different paths depending on what the contact says. Logs make those paths easier to review. Immediately after, the guide continues: Before reviewing Agent Logs, it helps to understand how the information is organized. Image 1 of 19 What this coversAgent Logs are especially helpful because AI agents are not always linear. The same agent may take different paths depending on what the contact says. Logs make those paths easier to review. Next stepBefore reviewing Agent Logs, it helps to understand how the information is organized.
All 8 steps in this procedure
  1. What did the contact say?
  2. How did Conversation AI respond?
  3. Did the AI call the expected action or tool?
  4. What information was passed into an action?
  5. What did the action return?
  6. Why did the AI take longer to respond?
  7. Did the AI use the knowledge base?
  8. Did the AI hand over or end the conversation?

Important concepts

Before reviewing Agent Logs, it helps to understand how the information is organized.

Conversation

A conversation is the full interaction between a contact and Conversation AI. It may include multiple customer messages and multiple AI responses.

Example:

A contact asks about availability, shares their name, asks for appointment slots, and later says they will confirm later. That full exchange is one conversation.

Conversation (image 2 of 19) What this shows A contact asks about availability, shares their name, asks for appointment slots, and later says they will confirm later. What this shows Illustrates the "Conversation" section of "Conversation AI - Agent Logs". This screenshot appears in the "Conversation" section of "Conversation AI - Agent Logs". The text alongside this image reads: A contact asks about availability, shares their name, asks for appointment slots, and later says they will confirm later. That full exchange is one conversation. Immediately after, the guide continues: A turn is one back-and-forth exchange inside a conversation. Image 2 of 19 How to confirmA contact asks about availability, shares their name, asks for appointment slots, and later says they will confirm later. That full exchange is one conversation. Next stepA turn is one back-and-forth exchange inside a conversation.

Turn

A turn is one back-and-forth exchange inside a conversation.

Example:

  1. The contact sends a message.

  2. Conversation AI processes the message.

  3. Conversation AI sends a response.

That is one turn. A conversation can have multiple turns.

Turn (image 3 of 19) What this shows That is one turn. What this shows Illustrates the "Turn" section of "Conversation AI - Agent Logs". This screenshot appears in the "Turn" section of "Conversation AI - Agent Logs". The text alongside this image reads: That is one turn. A conversation can have multiple turns. Immediately after, the guide continues: A step is an individual action that happens inside a turn. Image 3 of 19 What this coversThat is one turn. A conversation can have multiple turns. Next stepA step is an individual action that happens inside a turn.
All 3 steps in this procedure
  1. The contact sends a message.
  2. Conversation AI processes the message.
  3. Conversation AI sends a response.

Step

A step is an individual action that happens inside a turn.

A single turn may include steps such as:

  • User message received

  • AI agent invoked

  • Tool or action selected

  • Calendar availability checked

  • Knowledge base searched

  • Contact information extracted

  • AI response generated

  • Conversation ended

  • Human handover triggered

Step (image 4 of 19) What this shows User message received AI agent invoked Tool or action selected Calendar availability checked Knowledge base searched Contact information… What this shows Illustrates the "Step" section of "Conversation AI - Agent Logs". This screenshot appears in the "Step" section of "Conversation AI - Agent Logs". The text alongside this image reads: User message received AI agent invoked Tool or action selected Calendar availability checked Knowledge base searched Contact information extracted AI response generated Conversation ended Human handover triggered. This part of the guide covers 9 fields, listed below. Immediately after, the guide continues: Conversation AI Agent Logs can be accessed from three main areas. Image 4 of 19 What this coversUser message received AI agent invoked Tool or action selected Calendar availability checked Knowledge base searched Contact information extracted AI response generated Conversation ended Human handover triggered
Fields in this part of the guide User message receivedAI agent invokedTool or action selectedCalendar availability checkedKnowledge base searchedContact information extractedAI response generatedConversation endedHuman handover triggered
Next stepConversation AI Agent Logs can be accessed from three main areas.
All 9 steps in this procedure
  1. User message received
  2. AI agent invoked
  3. Tool or action selected
  4. Calendar availability checked
  5. Knowledge base searched
  6. Contact information extracted
  7. AI response generated
  8. Conversation ended
  9. Human handover triggered

Where to Access Conversation AI Agent Logs

Conversation AI Agent Logs can be accessed from three main areas.

1. Agent Logs Tab Under AI Agents (Universal View)

The main Agent Logs area gives users a central place to review Conversation AI sessions.

Users can go to AI Agents > Agent Logs to view the Agent Logs page.

From this page, users can review:

  • Sessions

  • Contacts

  • Metrics

  • Filters

  • Search results

  • Log details

1. Agent Logs Tab Under AI Agents (Universal View) (image 5 of 19) What this shows Sessions Contacts Metrics Filters Search results Log details What this shows Illustrates the "1. Agent Logs Tab Under AI Agents (Universal View)" section of "Conversation AI - Agent Logs". This screenshot appears in the "1. Agent Logs Tab Under AI Agents (Universal View)" section of "Conversation AI - Agent Logs". The text alongside this image reads: Sessions Contacts Metrics Filters Search results Log details. This part of the guide covers 5 fields, listed below. Immediately after, the guide continues: Conversation AI Agent Logs are also available from the Conversations page. Image 5 of 19 What this coversSessions Contacts Metrics Filters Search results Log details
Fields in this part of the guide SessionsContactsMetricsFiltersLog details
Buttons and menus referenced AI Agents > Agent Logs
Next stepConversation AI Agent Logs are also available from the Conversations page.
All 6 steps in this procedure
  1. Sessions
  2. Contacts
  3. Metrics
  4. Filters
  5. Search results
  6. Log details

2. Conversations Page

Conversation AI Agent Logs are also available from the Conversations page.

When a conversation includes Conversation AI activity, users can open the related log directly from the conversation thread. This helps users review the AI activity without leaving the inbox.

This is useful when a user is already reading a customer conversation and wants to understand how Conversation AI handled a specific message.

2. Conversations Page (image 6 of 19) What this shows This is useful when a user is already reading a customer conversation and wants to understand how Conversation AI handled a specific message. What this shows Illustrates the "2. Conversations Page" section of "Conversation AI - Agent Logs". This screenshot appears in the "2. Conversations Page" section of "Conversation AI - Agent Logs". The text alongside this image reads: This is useful when a user is already reading a customer conversation and wants to understand how Conversation AI handled a specific message. Immediately after, the guide continues: Agent Logs are available from contact records and contact side panels. This allows users to start from a contact and review the Conversation AI activity connected to that contact. Image 6 of 19 What this coversThis is useful when a user is already reading a customer conversation and wants to understand how Conversation AI handled a specific message. Next stepAgent Logs are available from contact records and contact side panels. This allows users to start from a contact and review the Conversation AI activity connected to that contact.

3. Contact Records and Contact Side Panels

Agent Logs are available from contact records and contact side panels. This allows users to start from a contact and review the Conversation AI activity connected to that contact.

This is helpful when the user knows which contact they want to investigate but does not know the exact session or log.

Agent Logs also work with contact side panel layout customization. Users can place the Agent Logs panel in supported side panel layouts, such as the center or right panel, based on their workspace preference.

3. Contact Records and Contact Side Panels (image 7 of 19) What this shows Agent Logs also work with contact side panel layout customization. What this shows Illustrates the "3. Contact Records and Contact Side Panels" section of "Conversation AI - Agent Logs". This screenshot appears in the "3. Contact Records and Contact Side Panels" section of "Conversation AI - Agent Logs". The text alongside this image reads: Agent Logs also work with contact side panel layout customization. Users can place the Agent Logs panel in supported side panel layouts, such as the center or right panel, based on their workspace preference. Immediately after, the guide continues: The Sessions view lists Conversation AI activity as individual sessions. Each row represents an AI session that can be opened for more detail. Image 7 of 19 What this coversAgent Logs also work with contact side panel layout customization. Users can place the Agent Logs panel in supported side panel layouts, such as the center or right panel, based on their workspace preference. Next stepThe Sessions view lists Conversation AI activity as individual sessions. Each row represents an AI session that can be opened for more detail.

Sessions View in Agent Logs

The Sessions view lists Conversation AI activity as individual sessions. Each row represents an AI session that can be opened for more detail.

Users can use the Sessions view when they want to review a specific interaction or find a log based on filters such as time range, contact, agent, or channel.

The Sessions view may show details such as:

  • Agent name

  • Contact name

  • Channel

  • Timestamp

  • Status

  • Action menu

Users can open a session to view the raw conversation and the execution timeline for that session.

Sessions View in Agent Logs (image 8 of 19) What this shows Users can open a session to view the raw conversation and the execution timeline for that session. What this shows Illustrates the "Sessions View in Agent Logs" section of "Conversation AI - Agent Logs". This screenshot appears in the "Sessions View in Agent Logs" section of "Conversation AI - Agent Logs". The text alongside this image reads: Users can open a session to view the raw conversation and the execution timeline for that session. This part of the guide covers 6 fields, listed below. Immediately after, the guide continues: Filters help users narrow down the logs and find the exact Conversation AI activity they want to review. Image 8 of 19 What this coversUsers can open a session to view the raw conversation and the execution timeline for that session.
Fields in this part of the guide Agent nameContact nameChannelTimestampStatusAction menu
Next stepFilters help users narrow down the logs and find the exact Conversation AI activity they want to review.
Example values
Agent name
Priya Raman
Contact name
Priya Raman

Illustrative values showing the expected format. They are not read from the screenshot.

All 6 steps in this procedure
  1. Agent name
  2. Contact name
  3. Channel
  4. Timestamp
  5. Status
  6. Action menu

Filtering Sessions

Filters help users narrow down the logs and find the exact Conversation AI activity they want to review.

Users may filter by details such as:

  • AI product

  • Channel

  • Contact

  • Time range

  • Agent

For example, a user may filter to show only Conversation AI sessions from live chat for a specific contact. This helps reduce noise when there are many sessions in the account.

Filtering Sessions (image 9 of 19) What this shows For example, a user may filter to show only Conversation AI sessions from live chat for a specific contact. What this shows Illustrates the "Filtering Sessions" section of "Conversation AI - Agent Logs". This animation appears in the "Filtering Sessions" section of "Conversation AI - Agent Logs". The text alongside this image reads: For example, a user may filter to show only Conversation AI sessions from live chat for a specific contact. This helps reduce noise when there are many sessions in the account. This part of the guide covers 5 fields, listed below. Immediately after, the guide continues: The Contacts view organizes Conversation AI activity around contacts instead of only sessions. Image 9 of 19 What this coversFor example, a user may filter to show only Conversation AI sessions from live chat for a specific contact. This helps reduce noise when there are many sessions in the account.
Fields in this part of the guide AI productChannelContactTime rangeAgent
Next stepThe Contacts view organizes Conversation AI activity around contacts instead of only sessions.
All 5 steps in this procedure
  1. AI product
  2. Channel
  3. Contact
  4. Time range
  5. Agent

Contacts View in Agent Logs

The Contacts view organizes Conversation AI activity around contacts instead of only sessions.

This is helpful when a user wants to start from a contact name and review the AI activity connected to that person.

For example, if a customer named Alex contacted the business multiple times, the Contacts view can help the user find Alex and open the related Conversation AI sessions.

The Contacts view is useful for:

  • Reviewing activity for a specific contact

  • Investigating repeated customer interactions

  • Understanding what Conversation AI did for one person

  • Finding sessions when the user does not know the exact session ID or timestamp

Contacts View in Agent Logs (image 10 of 19) What this shows Reviewing activity for a specific contact Investigating repeated customer interactions Understanding what Conversation AI did for one… What this shows Illustrates the "Contacts View in Agent Logs" section of "Conversation AI - Agent Logs". This animation appears in the "Contacts View in Agent Logs" section of "Conversation AI - Agent Logs". The text alongside this image reads: Reviewing activity for a specific contact Investigating repeated customer interactions Understanding what Conversation AI did for one person Finding sessions when the user does not know the exact session ID or timestamp. This part of the guide covers 3 fields, listed below. Immediately after, the guide continues: The Metrics tab gives users an overview of Conversation AI activity and performance. Image 10 of 19 What this coversReviewing activity for a specific contact Investigating repeated customer interactions Understanding what Conversation AI did for one person Finding sessions when the user does not know the exact session ID or timestamp
Fields in this part of the guide Reviewing activity for a specific contactInvestigating repeated customer interactionsUnderstanding what Conversation AI did for one person
Next stepThe Metrics tab gives users an overview of Conversation AI activity and performance.
All 4 steps in this procedure
  1. Reviewing activity for a specific contact
  2. Investigating repeated customer interactions
  3. Understanding what Conversation AI did for one person
  4. Finding sessions when the user does not know the exact session ID or timestamp

Metrics in Agent Logs

The Metrics tab gives users an overview of Conversation AI activity and performance.

While session logs help users understand a specific interaction, Metrics help users understand activity across multiple conversations.

Metrics may include:

  • Conversations handled

  • Contacts reached

  • AI messages

  • Average messages per conversation

  • Average response time

  • Top actions

  • Top agents

  • Top channels

  • Most active contacts

  • Conversation activity over time

  • Busiest hours

  • Channel usage over time

Users can filter Metrics by time range, agent, channel, and contact to focus on the data that matters to them.

Metrics in Agent Logs (image 11 of 19) What this shows Users can filter Metrics by time range, agent, channel, and contact to focus on the data that matters to them. What this shows Illustrates the "Metrics in Agent Logs" section of "Conversation AI - Agent Logs". This animation appears in the "Metrics in Agent Logs" section of "Conversation AI - Agent Logs". The text alongside this image reads: Users can filter Metrics by time range, agent, channel, and contact to focus on the data that matters to them. This part of the guide covers 12 fields, listed below. Immediately after, the guide continues: The Metrics view supports layout customization. Users can organize widgets based on what they want to monitor most often. Image 11 of 19 What this coversUsers can filter Metrics by time range, agent, channel, and contact to focus on the data that matters to them.
Fields in this part of the guide Conversations handledContacts reachedAI messagesAverage messages per conversationAverage response timeTop actionsTop agentsTop channelsMost active contactsConversation activity over timeBusiest hoursChannel usage over time
Next stepThe Metrics view supports layout customization. Users can organize widgets based on what they want to monitor most often.
All 12 steps in this procedure
  1. Conversations handled
  2. Contacts reached
  3. AI messages
  4. Average messages per conversation
  5. Average response time
  6. Top actions
  7. Top agents
  8. Top channels
  9. Most active contacts
  10. Conversation activity over time
  11. Busiest hours
  12. Channel usage over time

Customizing Metrics Layouts

The Metrics view supports layout customization. Users can organize widgets based on what they want to monitor most often.

For example, one user may want to focus on top actions and average response time, while another may care more about active contacts and channel usage.

Users can customize layouts so the most relevant information is easier to review.

Customizing Metrics Layouts (image 12 of 19) What this shows Users can customize layouts so the most relevant information is easier to review. What this shows Illustrates the "Customizing Metrics Layouts" section of "Conversation AI - Agent Logs". This animation appears in the "Customizing Metrics Layouts" section of "Conversation AI - Agent Logs". The text alongside this image reads: Users can customize layouts so the most relevant information is easier to review. Immediately after, the guide continues: When a user opens a session, the log detail view shows the conversation and the execution timeline. Image 12 of 19 What this coversUsers can customize layouts so the most relevant information is easier to review. Next stepWhen a user opens a session, the log detail view shows the conversation and the execution timeline.

Log Detail View

When a user opens a session, the log detail view shows the conversation and the execution timeline.

The log detail view generally includes:

  • The raw customer conversation

  • The execution timeline

  • User messages

  • AI responses

  • Tool or action calls

  • Inputs and outputs for selected steps

  • Latency for steps or turns

  • Status information

  • Copy or expand options where available

  • A visual graph of the execution path where available

This view helps users understand how Conversation AI moved from the customer message to the final response.

Log Detail View (image 13 of 19) What this shows This view helps users understand how Conversation AI moved from the customer message to the final response. What this shows Illustrates the "Log Detail View" section of "Conversation AI - Agent Logs". This animation appears in the "Log Detail View" section of "Conversation AI - Agent Logs". The text alongside this image reads: This view helps users understand how Conversation AI moved from the customer message to the final response. This part of the guide covers 9 fields, listed below. Immediately after, the guide continues: In the log detail view, users can review two important areas: the raw conversation and the execution timeline. Image 13 of 19 What this coversThis view helps users understand how Conversation AI moved from the customer message to the final response.
Fields in this part of the guide The raw customer conversationThe execution timelineUser messagesAI responsesTool or action callsInputs and outputs for selected stepsLatency for steps or turnsStatus informationA visual graph of the execution path where available
Next stepIn the log detail view, users can review two important areas: the raw conversation and the execution timeline.
All 10 steps in this procedure
  1. The raw customer conversation
  2. The execution timeline
  3. User messages
  4. AI responses
  5. Tool or action calls
  6. Inputs and outputs for selected steps
  7. Latency for steps or turns
  8. Status information
  9. Copy or expand options where available
  10. A visual graph of the execution path where available

Raw Conversation and Execution Timeline

In the log detail view, users can review two important areas: the raw conversation and the execution timeline.

Raw conversation

The raw conversation shows the messages exchanged between the contact and Conversation AI. This helps users confirm what the contact said and what Conversation AI replied.

Raw conversation (image 14 of 19) What this shows The raw conversation shows the messages exchanged between the contact and Conversation AI. What this shows Illustrates the "Raw conversation" section of "Conversation AI - Agent Logs". This screenshot appears in the "Raw conversation" section of "Conversation AI - Agent Logs". The text alongside this image reads: The raw conversation shows the messages exchanged between the contact and Conversation AI. This helps users confirm what the contact said and what Conversation AI replied. Immediately after, the guide continues: The execution timeline shows what happened behind the scenes. It displays the steps Conversation AI followed during the interaction. Image 14 of 19 How to confirmThe raw conversation shows the messages exchanged between the contact and Conversation AI. This helps users confirm what the contact said and what Conversation AI replied. Next stepThe execution timeline shows what happened behind the scenes. It displays the steps Conversation AI followed during the interaction.

Execution timeline

The execution timeline shows what happened behind the scenes. It displays the steps Conversation AI followed during the interaction.

For example, if a contact asks for appointment slots, the timeline may show that Conversation AI:

  1. Received the customer message

  2. Invoked the AI agent

  3. Selected the calendar availability action

  4. Sent the required details into the action

  5. Received available slots from the action

  6. Generated a response using the returned slots

  7. Sent the response to the contact

Execution timeline (image 15 of 19) What this shows Received the customer message Invoked the AI agent Selected the calendar availability action Sent the required details into the action… What this shows Illustrates the "Execution timeline" section of "Conversation AI - Agent Logs". This screenshot appears in the "Execution timeline" section of "Conversation AI - Agent Logs". The text alongside this image reads: Received the customer message Invoked the AI agent Selected the calendar availability action Sent the required details into the action Received available slots from the action Generated a response using the returned slots Sent the response to the contact. This part of the guide covers 7 fields, listed below. Immediately after, the guide continues: This helps users understand both the customer-facing conversation and the behind-the-scenes AI activity. Image 15 of 19 What this coversReceived the customer message Invoked the AI agent Selected the calendar availability action Sent the required details into the action Received available slots from the action Generated a response using the returned slots Sent the response to the contact
Fields in this part of the guide Received the customer messageInvoked the AI agentSelected the calendar availability actionSent the required details into the actionReceived available slots from the actionGenerated a response using the returned slotsSent the response to the contact
Next stepThis helps users understand both the customer-facing conversation and the behind-the-scenes AI activity.
All 7 steps in this procedure
  1. Received the customer message
  2. Invoked the AI agent
  3. Selected the calendar availability action
  4. Sent the required details into the action
  5. Received available slots from the action
  6. Generated a response using the returned slots
  7. Sent the response to the contact

This helps users understand both the customer-facing conversation and the behind-the-scenes AI activity.

Reviewing a Specific AI Message

Users can click a message or related log entry to review what happened for that part of the conversation.

For example, if a customer asks, “Can I book an appointment tomorrow?”, the related log may show whether Conversation AI called the calendar availability action, what details were sent to the action, and what result came back.

This helps users confirm whether the AI took the expected path for that customer request.

Reviewing a Specific AI Message (image 16 of 19) What this shows This helps users confirm whether the AI took the expected path for that customer request. What this shows Illustrates the "Reviewing a Specific AI Message" section of "Conversation AI - Agent Logs". This animation appears in the "Reviewing a Specific AI Message" section of "Conversation AI - Agent Logs". The text alongside this image reads: This helps users confirm whether the AI took the expected path for that customer request. Immediately after, the guide continues: Conversation AI can use tools or actions to complete tasks. These may include checking availability, booking appointments, extracting contact information, ending a conversation, or handing over to a team member. Image 16 of 19 How to confirmThis helps users confirm whether the AI took the expected path for that customer request. Next stepConversation AI can use tools or actions to complete tasks. These may include checking availability, booking appointments, extracting contact information, ending a conversation, or handing over to a team member.

Tool and Action Calls

Conversation AI can use tools or actions to complete tasks. These may include checking availability, booking appointments, extracting contact information, ending a conversation, or handing over to a team member.

When Conversation AI calls a tool or action, Agent Logs can show information such as:

  • Which tool or action was selected

  • What details were passed into the tool

  • What the tool returned

  • How the result was used in the AI response

  • How long the step took

This is useful because it helps users understand whether Conversation AI used the expected action for the customer request.

Example:

If a customer asks for appointment slots, the log may show a calendar availability action. The action input may include the requested date range, and the output may include the available slots returned by the calendar.

Tool and Action Calls (image 17 of 19) What this shows If a customer asks for appointment slots, the log may show a calendar availability action. What this shows Illustrates the "Tool and Action Calls" section of "Conversation AI - Agent Logs". This screenshot appears in the "Tool and Action Calls" section of "Conversation AI - Agent Logs". The text alongside this image reads: If a customer asks for appointment slots, the log may show a calendar availability action. The action input may include the requested date range, and the output may include the available slots returned by the calendar. This part of the guide covers 5 fields, listed below. Immediately after, the guide continues: Some log details may be available in both parsed and raw formats. Image 17 of 19 What to fill inIf a customer asks for appointment slots, the log may show a calendar availability action. The action input may include the requested date range, and the output may include the available slots returned by the calendar.
Fields in this part of the guide Which tool or action was selectedWhat details were passed into the toolWhat the tool returnedHow the result was used in the AI responseHow long the step took
Next stepSome log details may be available in both parsed and raw formats.
All 5 steps in this procedure
  1. Which tool or action was selected
  2. What details were passed into the tool
  3. What the tool returned
  4. How the result was used in the AI response
  5. How long the step took

Parsed and Raw Views

Some log details may be available in both parsed and raw formats.

Parsed view

The parsed view presents information in a more readable table-like format. This is easier for most users to understand because the data is organized into fields and values.

Raw view

The raw view shows the underlying structured data. This may be useful for advanced users who want to inspect the exact data passed between the AI agent and the selected tool or action.

For most users, the parsed view will be easier to review.

Raw view (image 18 of 19) What this shows For most users, the parsed view will be easier to review. What this shows Illustrates the "Raw view" section of "Conversation AI - Agent Logs". This animation appears in the "Raw view" section of "Conversation AI - Agent Logs". The text alongside this image reads: For most users, the parsed view will be easier to review. Immediately after, the guide continues: Agent Logs can show how long different steps took. This helps users understand why an AI response may have taken longer than expected. Image 18 of 19 What this coversFor most users, the parsed view will be easier to review. Next stepAgent Logs can show how long different steps took. This helps users understand why an AI response may have taken longer than expected.

Latency and Response Timing

Agent Logs can show how long different steps took. This helps users understand why an AI response may have taken longer than expected.

For example, if Conversation AI checks calendar availability and also extracts contact information, the response may take longer because multiple steps happened before the AI replied.

Latency can help answer questions such as:

  • Which step took the longest?

  • Did a tool call add time to the response?

  • Did the AI perform multiple actions before replying?

  • Was the delay related to a specific part of the interaction?

Latency and Response Timing (image 19 of 19) What this shows Which step took the longest? What this shows Illustrates the "Latency and Response Timing" section of "Conversation AI - Agent Logs". This screenshot appears in the "Latency and Response Timing" section of "Conversation AI - Agent Logs". The text alongside this image reads: Which step took the longest? Did a tool call add time to the response? Did the AI perform multiple actions before replying? Was the delay related to a specific part of the interaction? Immediately after, the guide continues: Q: Are Agent Logs meant only for troubleshooting? Image 19 of 19 What this coversWhich step took the longest? Did a tool call add time to the response? Did the AI perform multiple actions before replying? Was the delay related to a specific part of the interaction? Next stepQ: Are Agent Logs meant only for troubleshooting?
All 4 steps in this procedure
  1. Which step took the longest?
  2. Did a tool call add time to the response?
  3. Did the AI perform multiple actions before replying?
  4. Was the delay related to a specific part of the interaction?

Frequently Asked Questions

Q: Are Agent Logs meant only for troubleshooting?

No. Agent Logs are mainly a transparency view. They help users understand what Conversation AI did and how it handled a conversation. They can also help users investigate unexpected behavior.

Q: What is the difference between a conversation, a turn, and a step?

A conversation is the full interaction with a contact. A turn is one back-and-forth exchange between the contact and Conversation AI. A step is an action inside a turn, such as receiving a message, calling a tool, checking availability, or generating a response.

Q: What can users see in a log?

Users can review the customer message, AI response, execution timeline, selected tools or actions, inputs and outputs, latency, status, and related session details.

Q: What is the Sessions view?

The Sessions view lists Conversation AI sessions. Users can filter and open sessions to review the conversation and execution timeline.

Q: What is the Contacts view?

The Contacts view organizes logs by contact. This helps users start from a contact record and open the Conversation AI sessions connected to that person.

Q: What is the Metrics view?

The Metrics view shows aggregated Conversation AI activity, such as conversations handled, contacts reached, AI messages, average response time, top actions, top agents, top channels, and most active contacts.

Q: What is the difference between parsed and raw views?

The parsed view presents log data in a readable table format. The raw view shows the underlying structured data. Most users will find the parsed view easier to understand.

Q: Can users open Agent Logs without leaving Conversations?

Yes. When available, users can open the related Conversation AI Agent Log directly from the conversation thread.

Q: Do Agent Logs work with contact side panel layouts?

Yes. Agent Logs work with supported contact side panel layouts, allowing users to keep logs visible alongside contact information.

Was this guide helpful?

Related guides

Bot Status for Individual Contacts4 min read · 6 imagesGuided Form Based Setup for Conversation AI5 min read · 12 imagesConversation AI Agents Dashboard5 min read · 5 imagesConversation AI- Bot Snapshots4 min read · 3 images