Module 12: Triggers, Scheduling, and Interfaces
Duration: 60 minutes Day: Day 3, Session 6
The hands-on lab for this module lives in the Hermes repository. Read the Concepts page to understand the four interface patterns and why each serves a different operational need. Then open the Hermes lab guide to configure cron schedules and webhooks.
What This Module Is About
Your agent works when you run it manually. Now it needs to work when you are not there.
Automated operation requires two things: trigger mechanisms (how does the agent know when to run?) and interface patterns (how do humans and systems interact with the running agent?). This module covers both.
By the end of the lab, your agent will run on a daily schedule and respond to simulated webhook events — moving from "a tool you invoke" to "an operational service."
Learning Objectives
By the end of this module, you will be able to:
- Configure cron schedules for periodic agent tasks using Hermes's job scheduling system
- Set up webhook-triggered workflows that activate your agent in response to external events (alerts, deployments, pipeline completions)
- Explain four interface patterns — CLI, Slack/Teams, Webhooks, Cron — and select the right one for different operational scenarios
- Describe the Mission Control concept as the dashboard interface that consolidates agent activity across a fleet
Prerequisites
- Modules 10-11 completed (you have a working agent, ideally a fleet)
- Hermes running with cron and webhook support enabled
- Basic understanding of HTTP webhooks (if you've used Slack or GitHub webhooks, you're ready)
Module Contents
| Section | Content | Time |
|---|---|---|
| Reading | Concepts: Interface Patterns and Trigger Design | 15 min |
| Reading | Reference: Cron Config, Webhook Setup, Slack Integration | 10 min |
| Lab | Configure Triggers and Interfaces (Hermes repo) | 25 min |
| Quiz | Module 12 Assessment | 10 min |
| Exploratory | Stretch Projects | Optional |
Key Insight: Interfaces Define When Agents Are Useful
A manually-invoked agent is useful when you remember to run it. An automatically-triggered agent is useful when you need it — regardless of whether you were thinking about it. The difference between these two is the difference between a tool and an operational service.
The interface pattern you choose determines the agent's operational profile: cron for predictable periodic work, webhooks for event-driven response, chat for conversational on-demand access, dashboards for situational awareness. Most production agents use multiple interfaces for different interaction modes.