Workflow Builder | Product Updates | Mithrilis

Overview#

The Workflow Builder replaces manual, repetitive logistics processes with visual automations that your team can build, test, and deploy — no coding required.

Drag and drop triggers, logic, and actions onto a canvas and wire them together into end-to-end workflows. Route carrier alerts to Slack, classify exceptions with AI, sync shipment data across systems — the builder handles it all.

Workflow Builder canvas with a live exception routing workflow

Building Workflows#

The canvas works like a whiteboard: pan, zoom, and drag steps around to organize your workflow visually.

Step Library#

The left sidebar has a searchable library of all available steps, organized into two categories:

Triggers — Events that start your workflow (carrier updates, scheduled checks, incoming emails)
Actions — Steps that do the work (send Slack messages, call APIs, query your data, run AI models)

Drag a step onto the canvas, then connect steps together to define how data flows through your automation.

Test Individual Steps#

You can test any step on its own without running the full workflow. Click a step to run it — it turns green on success and red on failure, with results shown inline so you can verify the output before connecting the next step.

Test as you build

Run individual steps frequently as you build your workflow. This catches issues early and lets you verify that each step produces the data you expect before wiring everything together.

50+ Integrations#

Every integration step includes the provider's logo, making it easy to identify what each step does at a glance. The canvas stays clean and readable no matter how complex your workflow gets.

Step library with 50+ available integrations organized by category

Triggers#

Every workflow starts with a trigger — the event that kicks it off.

Carrier and System Updates#

Connect your carrier APIs, TMS, or any external system to Mithrilis. When an event happens — like a shipment status change — the workflow runs automatically.

Example: A carrier sends a tracking update → the workflow checks if the shipment is late → alerts your team on Slack.

Scheduled Checks#

Set workflows to run on a recurring schedule — hourly, nightly, weekly, or any interval you need.

Example: Every night at 10 PM, check for shipments missing proof-of-delivery and create exceptions for any that are overdue.

Incoming Emails#

Point an email inbox at Mithrilis and the workflow runs when a new message arrives. The system can classify the email type automatically before processing it.

Example: Carrier rate confirmation emails are automatically parsed and the structured data is extracted into your system.

One trigger per workflow

Each workflow has one trigger. If you need the same automation to respond to multiple event types, create separate workflows for each and have them share steps.

Data Flow Between Steps#

Data flows automatically from one step to the next. When a step runs, its output is available to every step that comes after it.

For example, when a carrier update triggers a workflow, the AI classification step can reference the carrier name, shipment ID, and status from that update in its configuration. A built-in picker lets you browse and insert data from any previous step — you can see the actual values flowing through the workflow, not just field names.

Data picker showing outputs from previous steps with live data preview

AI-Powered Steps#

Add AI steps to bring intelligence into your automations. AI steps can classify, extract, summarize, and transform data — turning unstructured carrier communications into structured, actionable information.

Choose from multiple AI models and configure what the AI should do using plain language instructions. You can have the AI return structured data that feeds directly into the next step of your workflow.

Example: Classify incoming carrier exceptions by severity and category, then route critical ones to Slack while batching lower-priority ones into a daily digest email.

Branching Logic#

Route data down different paths based on conditions. Define multiple branches — the workflow follows the first one that matches, with a default path for everything else.

Example: Route exceptions to different Slack channels based on severity:

Critical → #ops-critical
High → #ops-alerts
Everything else → #ops-general

Version History and Deployment#

Safe Iteration#

Every save creates a new version, and you can browse the full history to see how a workflow has evolved. Undo and redo are fully supported.

Draft vs. Live#

Workflows have two states:

Draft — Only runs when you click Test. Safe for experimentation.
Live — Runs automatically when the trigger fires.

This means you can iterate on a workflow without affecting anything in production. When you're ready, publish the latest draft to make it live.

Templates#

Start from pre-built templates for common patterns like carrier alert routing, exception triage, and daily digest emails. Customize them to fit your operations.

Test before publishing

Always test your workflow with realistic data before publishing. Use the Test button to verify every branch and make sure exceptions are routed correctly.

Error Handling#

When a workflow step fails, it doesn't just disappear. Failed runs are captured with full details — you can see which step failed and why, then replay the run after fixing the issue.

You're notified about failures through in-app alerts, error badges on the workflow, and email notifications so your team can respond quickly.

Getting Started#

Create a new workflow

Click New Workflow from the Workflows page. Give it a descriptive name (e.g., "Late Pickup Alert") and choose a trigger type.

Set up the trigger

Configure how the workflow starts. Connect it to a carrier or TMS for real-time updates, set a recurring schedule, or point it at an email inbox.

Add steps

Drag steps from the library onto the canvas. Connect them in the order you want data to flow, and use the data picker to reference information from previous steps.

Test and publish

Click Test Run to execute the workflow with sample data. Check the results at each step. When everything looks right, click Publish to make it live.