View Execution Log of Agent OS

The Execution Log section in Agent OS helps you monitor the status of your automations and agents.

Additional ResourceRefer to the Executions in Agent OS documentation to learn about the execution statuses.

To access Execution Log, log into your Contentstack account and perform the following steps:

1. After logging in, click the App Switcher icon, then select Agent OS from the list.Click to enlarge
2. Open your project or create a new one.
3. From the top navigation panel, click Settings.Click to enlarge
4. Click Execution Log in the left navigation panel. You can view the executions for Agents and Automations by clicking the dropdown.Click to enlarge

   On the listing page, you see the following columns:

   1. Name: Displays the name of the automation or agent that was executed.
   2. Status: Shows the final execution state (e.g., Success or Failed) or intermediate states (Running, Pending, Paused, and Rejected).
   3. Started At: Indicates the date and time when the execution began.
   4. Duration: Shows how long the execution took to complete.
   5. Connectors Used: Indicates which connectors or services were used during the execution.
   6. Tools Used: Indicates the tools used during the agent execution.

   NoteTo view the log, you must execute an automation or agent. 

   Click to enlarge

5. To view more details of a particular execution, click a specific log. You will see a time-wise distribution of each step.

Execution steps

The Execution Steps timeline shows every action performed during the run in chronological order, making it easy to trace agent behavior and identify performance bottlenecks.

Agent execution detail includes:

• Step-by-step execution flow (for example, web search, content creation, message delivery)
• Metrics:
  • Started At: Shows the exact date and time when the execution began.
  • Duration: Indicates how long the execution took to complete from start to finish.
  • AI Provider: Shows the provider of the AI model used for the run. Agents that use the Contentstack-managed model show Contentstack Managed.
  • Total Tokens: Represents the total number of tokens consumed during the run.
  • Model: Identifies the AI model used to execute the task.

  Note Total Tokens and Model appear only when Bring Your Own Key (BYOK) is enabled for your organization and the agent uses a custom LLM. For more information, refer to the Custom Models documentation.

• Input and Output:
  • Input:
    • Displays the exact prompt or instructions provided to the agent.
    • When files were sent to the agent through its HTTP trigger, lists each file under attachments with its name, size and type, for example "guide.pdf, 311 KB, application/pdf". File contents are not stored in the log.
    • Shown in JSON format for reproducibility.
  • Output:
    • Displays structured execution results in JSON format.
  Click to enlarge
• Files created by the agent: When the agent creates a file, the run shows a Create A File step. Click it to see a file card with the file name, size and time until the link expires (for example, "Expires in 23h 59m"), plus buttons to copy the download link and to download the file. For more information, refer to the Files by Agent OS documentation.Click to enlarge

Automation execution detail includes:

• Step-by-step execution flow (for example, web search, content creation, message delivery)
• Metrics:
  • Started At: Shows the exact date and time when the execution began.
  • Duration: Indicates how long the execution took to complete from start to finish.
• Input and Output:
  • Input:
    • Displays the trigger payload or initial input data/configuration for the automation.
    • Shown in JSON format for reproducibility.
  • Output:
    • Displays structured execution results in JSON format.
  Click to enlarge