All articles

Monitor AI Agent Activity

Last updated on

Wrike Documentation Team

Wrike Documentation Team

TL;DR

Monitor deployed AI agents in the Agent Activity Dashboard to track actions, triggers, success/failure status, and AI reasoning. Use filters to debug runs and review logs weekly. To limit excessive actions, narrow agent scope with action filters, target specific item types, use triggered sub-items, and avoid placing agents on huge folders.

Table 10. Availability

Free Team Business Pinnacle Apex
Availability: Business, Pinnacle, Apex. ; Unavailability: Free, Team;

Overview

Once your AI agents are deployed, you can monitor their performance in the Agent Activity Dashboard within the AI agents management interface.

AI Agent Overview Table

See all your appointed AI Agents with details including:

  • Agent name and monthly activity summary (e.g., "5 actions this month").
  • Recent action timestamps.
  • Action types (such as changing custom fields or posting comments).
  • Work items affected.
  • Trigger events that activated the agent.
  • Success or failure status.

Group by runs: When a single trigger fires an agent that takes several actions, you can group those actions together as one run, so the log shows one entry per execution instead of one row per action. This makes it easier to see what the agent did in response to a single event.

Filter AI Agent Activity Logs

Use the filter controls above the activity table to narrow the log and locate specific runs faster.

Filters available: Timestamp, Action, Work item, Triggered by, Status.

Useful in high-volume spaces where the agent has run hundreds of times - debug a failure by filtering to the failing status, or trace a specific work item's history.

Detailed Action Views

Click any action for more information, including:

  • Complete agent details and timestamp.
  • Specific work item and its location.
  • Exact trigger that activated the agent.
  • Precise action taken (e.g., "Update Request Category field to creative asset").
  • Action name: If you named the action during setup, the name appears alongside the action type (e.g., "Set Priority – Change custom field").
  • AI reasoning with a full explanation of why the agent made its decision.

Monitoring Best Practices

  • Regular review: Check AI agent activity logs weekly to make sure agents are working as expected.
  • Reasoning analysis: Use the detailed reasoning display to understand decisions and spot opportunities to improve prompts.
  • Success rate tracking: Monitor success and failure statuses to catch ongoing issues.
  • Performance adjustment: Refine agent prompts based on activity logs to boost accuracy and consistency.

How to Limit the AI Agent from Taking Too Many Actions

Mind Your Scope - the most critical step in building effective automation.

  • Avoid Select All: Selecting all available scopes can lead to unintended actions and performance issues.
  • Use Action Filters to define exactly where an agent should operate.
  • Target Specific Items: Configure agents to only work on specific item types or statuses (e.g., "active").

Leverage Triggered Sub-items

  • Automate Project Blueprints: Set the agent to perform actions on all sub-items the moment the parent project's status changes.
  • Action Precision: Ensures automation flows downward through a project structure exactly when needed.

Manage Folder Capacity

  • Limit Item Counts: Do not place agents on folders with a massive number of items (e.g., 20,000 items).
  • Prevent Performance Trouble: Large folder volumes can cause technical issues for the agent's processing.

What’s Next?

0 comments

Please sign in to leave a comment.

Top

Related articles