Moxo AI agents deep-dive: Planner vs. Review vs. Support agents
Moxo AI Agents are intelligent workflow assistants embedded inside your workflows that automate repetitive, predictable tasks while humans remain in control of critical decisions.
They handle three distinct jobs: preparing actions before anyone touches them, reviewing submissions against your criteria, and providing real-time guidance when participants get stuck.
Think of them as the operations team you always wanted but could never hire. They prep, validate, route, and nudge at scale while you focus on the judgment calls that actually require your brain.
In this deep-dive, we'll break down the three core agent roles in Moxo, explain exactly how each one works, and show you when combining them creates workflows that practically run themselves (with you still in the driver's seat, of course).
Key takeaways
Moxo AI agents automate the heavy lifting around workflow steps. They prep tasks, validate inputs, and guide users while humans oversee the decisions that matter.
Each agent addresses a unique bottleneck. Planner handles preparation, Reviewer handles validation, and Support handles guidance. Different problems, different agents.
Combined effectively, these agents reduce manual work and speed throughput without compromising governance or creating accountability black holes.
Best practices matter. Confidence thresholds, escalation paths, and human-in-the-loop checkpoints determine whether AI agents help or create new problems.
What are Moxo AI agents
Moxo AI Agents are embedded process intelligence inside your flows. They prepare work, validate submissions, and guide participants through each step so your team isn't manually babysitting every action.
Here's the core idea: every workflow contains two types of work. There's the judgment work only humans can do (approvals, exceptions, risk calls).
And there's the execution work around those decisions (staging documents, checking if uploads meet requirements, answering the same questions for the fiftieth time). Moxo AI agents handle the second category. Humans stay accountable for the first.
Comparison: The three Moxo AI agents at a glance
The Planner: Auto-generating workspace tasks (AI Prepare Agent)
The Moxo Planner gets workflow actions ready before humans ever see them by setting titles, extracting data from previous steps, prefilling forms, and attaching required files.
You know the pain here. Someone triggers a workflow and the assignee receives a blank form. No context. No attached documents. No indication of what they're supposed to do with this thing. So they ping you. Or they guess. Or they do it wrong and you spend more time fixing it than if you'd just done it yourself.
How Moxo's Planner works. You assign the AI Prepare Agent as the preparer on a workflow action. It applies your configured logic, pulls context from the workspace, extracts known information from previous steps, and stages everything so the assignee receives a ready-to-go action instead of a mystery box.
Example use case. In a client onboarding workflow, the Moxo AI Prepare Agent pre-fills form fields from existing CRM data, attaches required PDFs, and sets clear instructions. Your team receives actions that are 80% complete instead of starting from scratch every time.
The Reviewer: Validating document uploads (AI Review Agent)
The Moxo Reviewer automates file verification against your defined requirements so you're not manually checking every upload for completeness.
Here's the scenario that drives ops teams insane: you request specific documents with specific formatting requirements. The client uploads a blurry photo of a document that's clearly from 2019. Or they send the right document but it's missing a signature. Or they attach a file called "final_FINAL_v3_USE_THIS_ONE.pdf" that is, in fact, not the right one at all. Someone has to check. That someone is usually you.
How Moxo's Reviewer works. Once a file is uploaded, the AI Review Agent evaluates it against your configured metadata and rules. If the submission meets criteria with high confidence, the action advances automatically. If confidence is low, the agent reopens the action with specific feedback so the participant knows exactly what to fix. Edge cases below your threshold escalate to a human reviewer.
The control is configurable. Set confidence thresholds based on risk (70-80% for typical checks, 90%+ for sensitive documents). Define escalation paths. You decide where the line is between "AI handles this" and "a human needs to look."
The Supporter: Real-time client assistance (AI support/chat agent)
The Moxo Supporter provides contextual help inside workflows so participants can answer their own questions instead of emailing you.
You've fielded this message a hundred times: "Quick question, how do I fill out this section?" Or: "What format do you need for the document?" Each question takes two minutes to answer. But you get forty of them a week. That's over an hour of your time just explaining the same things to different people.
How Moxo's Support Agent works. Users trigger the AI chat assistant by mentioning it or requesting help within the workflow. The agent pulls from workflow context and connected knowledge to provide accurate, relevant answers. It can explain form fields, clarify compliance requirements, and guide clients through onboarding steps. Complex queries escalate to a human.
When to combine all three Moxo agents
For most enterprise workflows, combining Moxo's Planner, Reviewer, and Support agents yields the greatest impact. Each addresses a different bottleneck, and together they cover the full lifecycle of workflow execution.
Example: Complex onboarding. The Moxo Planner auto-prepares tasks using CRM data. The Reviewer validates uploaded IDs and contracts against your requirements. The Support Agent answers real-time questions. Humans step in based on your escalation rules for approvals and exceptions.
AI handles the coordination. Humans handle the decisions. That's not a compromise. That's the model.
Best practices for using Moxo AI agents
Configure confidence thresholds wisely. Too low and you'll review everything manually anyway. Too high and legitimate submissions get flagged unnecessarily. Start at 75-80% for standard processes and adjust based on what you're seeing.
Audit logs help trace Moxo AI decisions. When someone asks why an action was approved or flagged, you need the receipts. Make sure your configuration supports the compliance requirements of your environment.
Keep humans in the loop for judgment-heavy steps. Moxo AI agents should never be the final word on high-stakes decisions. Design your workflows so approvals, exceptions, and risk calls route to humans automatically.
Combine Moxo agents based on workload types. Prep tasks first, review next, then support during execution. The sequence matters because each agent builds on what the previous one established.
Streamline business operations
Moxo AI Agents exist to automate the work around the work. The Planner stages actions so they arrive ready to execute. The Reviewer validates submissions so humans don't manually check every upload. The Support Agent answers questions so participants can progress without waiting.
None of these agents make decisions for you. They make it possible for decisions to happen at the right moment, with the right context, and without the manual coordination that usually slows everything down.
If you're already using Moxo, enabling these agents is the next step toward workflows that scale without proportional increases in manual effort. Pick your highest-volume process and enable the Planner first. The impact becomes obvious fast.
Ready to configure AI agents in your workflows? Get started today.
FAQs
What are Moxo AI agents and how do they differ from chatbots?
Moxo AI agents are process-aware assistants embedded inside workflows, not standalone conversational tools. They operate within the context of your specific process, handling preparation, validation, and guidance based on workflow rules you define. Chatbots respond to general queries. Moxo AI agents execute within structured business operations.
How does the Moxo Review Agent decide when to escalate?
The Review Agent uses configurable confidence thresholds you set during workflow configuration. When a submission scores below your threshold, it automatically reopens the action with feedback or routes to a designated human reviewer.
Can I use Moxo AI agents for processes involving external clients?
Yes. The Support Agent provides real-time, context-aware answers to external participants within the workflow experience. Clients can get guidance on requirements and next steps without emailing your team.
Do Moxo AI agents replace human oversight?
No. Moxo AI agents automate repetitive execution work while preserving human accountability for decisions. Approvals, exceptions, and risk calls still require human judgment.
How do I start using Moxo AI agents?
Enable agents in the Manage tab of your Moxo workspace, then configure them inside individual workflows with your defined rules, thresholds, and escalation criteria.
