✔️ AI Agents December Update: Status Actions & Agent Chaining

TL;DR: December’s Wrike Agent News unlocks true workflow automation! Agents can now set task statuses and trigger each other, enabling powerful multi-step workflows like auto-escalating stale approvals. Agents now read numerical custom fields (hours, deal value) for smarter routing and stay silent if no action is needed. Management just got easier with clickable deployment links and improved logs. Featured use case: a three-agent chain for automatic approval escalation saves managers hours weekly. Dive in for more use cases, setup guides, and reach out in comments if you want help configuring these new capabilities!

Hey Wrikers! 👋

Welcome to Wrike Agent News

This monthly digest features updates, learnings, and how-tos from the Wrike R&D team.

Thank you for your continued feedback as we build out AI Agents. The past thirty days, we shipped capabilities that unlock true workflow automation - Agents can now change statuses and trigger each other, enabling multi-step processes that previously required manual intervention.

🎯 New Action

Change Status Agents can now set task status - not just react to status changes, but actually move work forward. Combined with November's status change trigger, you now have complete status workflow automation.

Use case: Task sits in "Pending Approval" for 5+ days → agent changes status to "Escalated" and @mentions the department head.

Use case: All subtasks complete → agent automatically sets parent task to "Ready for Review".

🔗 Agent Chaining

Agents Trigger Agents: Build multi-step workflows where one agent's action triggers the next agent. No manual handoffs required.

Use case: Triaging Agent classifies request and sets category field → field change triggers Assignment Agent → assigns to specialist. Classification-to-assignment in seconds.

Automations Trigger Agents: Your existing Wrike automation rules can now kick off agent analysis. This means you can use simple rule-based triggers to launch intelligent AI processing.

Use case: Automation moves task to "Ready for Review" folder → triggers Quality Check Agent that analyzes deliverables against brief requirements.

🔢 Expanded Context

Numerical Custom Fields Agents can now read number, percentage, and currency fields - giving them quantitative context for smarter decisions.

Use case: Agent reads "Estimated Hours" field → flags requests over 40 hours for manager review before assignment.

Use case: Agent reads "Deal Value" currency field → routes high-value opportunities to senior team.

🔇 Silent When Appropriate

Agents can now choose not to act. If analysis concludes no intervention is needed, the agent exits silently—no empty comments, no notification spam.

Previously, agents always posted something when triggered. Now agents only speak up when there’s something to say.

🛠️ Management Improvements

Clickable Agent Locations: Jump directly from the agent management table to where an agent is deployed. No more hunting through folders.

Improved Logs Dashboard: The entire log row is now clickable to open details. Plus, empty states help you understand when agents haven't run yet versus when there's nothing to report.

Agent of the Month: Stale Task Escalation

In this section, we will dive into a specific use case.

Use case: Tasks get stuck. Approvals sit in queues. Reviews languish. Blocked items stay blocked. By the time anyone notices, it's a fire drill.

"Legal reviews can sit for weeks. We need automated escalation when approval time exceeds our SLA."

The Solution: A three-agent chain that demonstrates this month's new capabilities:

• Agent 1: Detects stale tasks and updates a custom field.
• Agent 2: Triggered by field change, escalates by changing status.
• Agent 3: Triggered by status change, posts comment @mentioning project manager.

Time Saved: 3-5 hours/week per manager.
Prerequisites: Custom field for "Stale" flag, workflow with escalation status .
Deployment Time: 25-30 minutes (three agents).

How It Works

This is agent chaining in action:

• Agent 1 (Stale Detector): Scheduled daily, scans for tasks stuck too long, sets "Stale" field to "Yes".
• Agent 2 (Escalator): Triggered by field change, changes status to "Escalated".
• Agent 3 (Notifier): Triggered by status change, posts comment @mentioning project manager.

Three simple agents. One complete workflow 💯

Complete Configuration

👉  Get the full Stale Task Escalation setup guide 

The complete guide includes:

• Both agent prompts with customization instructions.
• Step-by-step deployment for the two-agent chain.
• Examples for approval, creative, and IT workflows.
• Troubleshooting tips.

Combining This Month's Features

The real power comes from combining status change actions with agent chaining. The Stale Task Escalation above is one example. Here's another:

Complete Intake Pipeline:

• New task created → Triaging Agent classifies and sets category field.
• Field change triggers Assignment Agent → routes to appropriate specialist.
• If a task gets stuck, Stale Detector flags it → triggers Escalator → Notifier alerts PM.

Each agent does one job well. Chained together, they handle entire workflows without manual intervention.

Questions?

Want help setting up the Approval Escalation Agent?
Reach out to us in the comments.

Need a different agent type?
Contact your Customer Success Manager.

Found a bug or have feedback?
Use the feedback link in Wrike Labs.

Rohan V Community Team at Wrike Wrike Product Manager Become a Wrike expert with Wrike Discover

Rohan V Wrike Team member Become a Wrike expert with Wrike Discover