In the Monitoring tab, you can review every action that has been carried out by your automation and monitor them. The graph displays the various metrics of your automation pipelines, and it can also be filtered to highlight any specific information you’re looking for.

You can click on a pipeline to examine it using this tool.

On this chart, you can see the data of all actions that have taken place across all your pipelines in your team, displayed across a timeline. You can modify the time frame to look at the data in smaller or larger scales. You can also filter by pipeline to look at data only for specific pipelines.

Chart parameters

The monitoring chart can be given parameters to help you examine your pipelines in detail:

• Breakdown - You can view your chart’s data by Status, Action or Agent, Event, Project and Pipeline. Depending on the selection, you'll see the percentage of each status below the chart, with the total number of actions or events shown in the bottom right corner.
  • Status - You can view your actions by status, and you can select multiple at once.
  • Event - You can have the graph display action data based on the event that triggered it.
  • Action or Agent - You can select both default or custom actions to see their details on the graph. For custom actions, you can select them by name from the dropdown list.
  • Project - You can have the data display the projects in which your pipeline has been triggered.
  • Pipeline - You can have the data display the pipelines in which your actions have been triggered.
• Measure - You can show the data based on Count and Computation time.
• Time units - Aggregates the chart data by the selected unit:
  • Second - Available for windows under 5 minutes.
  • Minute - Available for windows under 3 hours.
  • Hour - Available for windows under 7 days.
  • Day - Available for windows under 90 days.
  • Month

Filter methods

There are a couple of methods you can use to filter both the Monitoring chart and the table below.

Date and Time

You can also filter the data by date and time using the date picker above the chart:

• Custom - Shows all data available within a range of dates that you select. This time frame can only be within the past six months.
• Today - Shows all data available from all hours of the current day.
• Yesterday - Shows all data available from all hours of the previous day.
• 7D - Shows all data available from the past seven days, with each day separated into 00:00 and 12:00 time blocks.
• 30D - Shows all data available from the past thirty days in 2-day increments.
• 6M - Shows all data available from the past six months.

The graph will show you data from Today by default.

You can also click and drag over bars (days, months) on the chart to select a specific and smaller time frame to zoom in on the details within the selected time.

Filters

You can use filters to narrow the actions shown in Monitoring by name, event, status, pipeline, or project. You can also filter Custom action and Send webhook actions by values in their arguments.

Click Reset filters at the top to clear your filters.

Add filter options

To add a filter option:

1. In Monitoring, click Filters +.
2. Select any filters you need under Name, Triggered by event, Status, Pipeline, and Project.
3. To filter by action arguments, click + Add Filter below Project, select Custom action or Send webhook, and configure the condition.
4. Click Apply.

Filter options

These are the filter options available, and what operators you can use for them:

• Name:
  • Option: You can select one or multiple options, either individually or by option type (default action, custom action, agent).
• Triggered by event:
  • Options: You can select one or multiple events, either individually or by event type.
• Status:
  • Options: You can select one, multiple, or all statuses.
• Pipeline:
  • Operator: is, is not.
  • Options: You can select one or multiple pipelines.
• Project:
  • Operator: is, is not, is empty.
  • Options: You can select one or multiple projects.

Filter by action arguments

You can filter actions by values in a custom action's input Event or Context, or in a Send Webhook action's Request or Response headers and payload. Each argument filter defines one condition.

After adding a Custom action or Send webhook filter:

1. Select the data to filter:
   • Custom action: Under Input source, select Event or Context. Event is selected by default.
   • Send webhook: Under Call, select Request or Response. Under Source, select Headers or Payload. Request and Headers are selected by default.
2. In JSON Path, enter the path to the field within the selected data. The $. prefix is provided automatically, so enter after.status for $.after.status. Leave the path empty to filter the selected data at the root level.
3. Select an Operator.
4. In Value, enter the value to compare as valid JSON. No value is required for Empty or Not empty.
5. Add more conditions as needed, then click Apply.

Operators and values

Text values require double quotes. Numbers and booleans do not. To enter a JSON null value, use null without quotes.

Date and time values are not treated as a separate type. Use number operators and an unquoted number for numeric values, or text operators and a quoted string for text values.

Empty matches null, "", {}, or []. Not empty matches non-empty values. Neither operator requires a value. Leaving Value blank is treated as matching empty or null values.

Matching behavior

Equals requires an exact match, while Contains checks whether the specified value is included in the field.

For an array field, Contains accepts a single value, such as 3, or an array, such as [1,3]. You do not need to enter the full array to check for specific elements.

Use Not equals or Not contains to exclude matches instead.

Example: To find custom actions whose Context contains after.status with the value InProgress, select Custom action and Context, enter after.status in JSON Path, select Equals, and enter "InProgress" in Value.

Combining conditions

Click + Add Filter to add another condition, then select AND or OR between conditions. AND is selected by default. Use AND when both conditions must match, or OR when either condition can match.

When combining three or more conditions, AND is evaluated before OR. For example, Condition 1 OR Condition 2 AND Condition 3 is evaluated as Condition 1 OR (Condition 2 AND Condition 3).

You can add up to 10 argument filters, including multiple filters for the same action type or a combination of Custom action and Send webhook filters. Each condition counts as a separate filter on the Filters button.

To remove a condition, click its x button. Other conditions remain unchanged.

Share a filtered view

The Monitoring URL includes your applied argument filters. Copy the URL to share or reopen the filtered view.

List of actions

Columns

Below the graph, you can see the information of every regular action or agent action laid out in a list with a handful of columns:

• Name - The name of the action or agent.
• Details - The action’s description.
• Pipeline - The pipeline in which the action was triggered.
• Project - The project in which the action was triggered.
• Triggered by Event - The name of the event triggering the action.
• Event Metadata ID - The ID of the project, folder, or item that triggered the action.
• Creation date - The date that the action was created.
• Status - The current status of the action.
• Actions - Column where you can rerun or revoke action.

There are various ways you can manage your columns:

• Resize - Click and drag the edge of any column to resize.
• Freeze - To freeze a column in a static position on the left side of the table, hover over the column and click the Freeze icon in the header.
• Move - Click and drag the header of any column to move and reorder it.
• Sort - You can click the up or down arrows on a column's header to sort its corresponding data by ascending or descending order, respectively. By default, the table is sorted by Creation Date.

If you Resize, Freeze, and Move your columns, the state of your Pipelines table will be saved for you. Any time you refresh or revisit the Pipelines table, it will maintain your latest changes.

From the Column menu ☰, you can select and deselect columns to toggle their visibility on the table. As well as this, you may create Custom Columns to display custom metrics that you can apply to projects.

Rerun and Revoke

If a previously executed action’s status is Success, Failed, or Revoked, you can click the Rerun button to rerun it with its original payload. It’ll also run all subsequent nodes of the action or agent. As an example, if the action had custom arguments added to the event object, the action will be rerun using its original custom arguments. If the agent’s prompt was edited, the agent will rerun using its original prompt.If the agent’s parameters were edited, the agent will rerun using its original parameters.

The only exception is that if the action’s code was edited or a secret updated, the action will be rerun using the new updated code and secret.

Rerun is unavailable for disabled pipelines, for deleted actions or agents, and for nodes removed from the pipeline.

If an action is Pending or Received, you can click the Revoke button to stop it from being carried out.

You can click on an action to open a drawer menu with all of its corresponding data. The details are dependent on either:

• the action type (Custom, Webhook, or Email)
• the agent

From the top right corner of this menu, you may choose to view the action in its corresponding pipeline. For custom actions, you may also open them in the code editor directly from the menu.

Action statuses

An action can have one of these statuses:

• Received - this shows when the action has been received by the system but it isn’t in progress yet.
• Pending - this shows when the action is in progress.
• Success - this shows when the action was carried out successfully.
• Failed - this shows when the action failed to be carried out.
• Revoked - this shows that the action was revoked before it could be carried out.

Agents

• Details - This tab shows the agent’s corresponding parameters, including the model prompt.
• Chain - The chain of triggers leading to the selected agent.
• Input - The prompt given by the user.
• Output - The model’s response, written in SuperAnnotate’s JSON format. If the model fails, the output will contain the error message in the same format. For advanced prompts, if the JSONPath has a missing value or the key doesn’t exist, the output will contain an error text message.
• Traces - This tab shows the tool calls that were made by the agent using the SuperAnnotate MCP tool. It displays the request and response payloads for each agent’s tool call.

Custom action

• Details - This tab shows the action’s corresponding data based on its configuration. This includes the action’s status, Team API key, Secret, custom arguments, description, and its creation, start, and end dates.
• Chain - This tab shows the chain of triggers leading to the selected action. If configured, the action’s Team API key is also shown here.
• Input - The event JSON and a context object.
• Output - The output JSON, text or traceback.
• Logs - The custom action’s logs, available so that you can closely monitor the automations and troubleshoot any given errors.

Webhook

• Details - This tab shows the action’s corresponding data based on its configuration. This includes the status, URL, secret, and its creation, start, and end dates.
• Chain - The chain of triggers leading to the selected action.
• Request - The corresponding request headers and payload.
• Response - The corresponding response headers and payload, as well as the status code.

Email

• Details - This tab shows the action’s corresponding data based on its configuration. This includes the status, email, and its creation, start, and end dates.
• Chain - The chain of triggers leading to the selected action.
• Email body - The event that took place, and what it was triggered by, will show in this tab.