Skip to main content
With Colossal Agents you can automate your commerce operations by describing what you want in the chat interface, or build visually with the editor. The Agent Builder composes triggers and actions from your installed apps and from Colossal App into agents.

How it works

An agent connects a trigger (an event that starts it) to a sequence of steps:
  1. A trigger fires: a commerce event from the Colossal app, a webhook from a third-party app, or a schedule
  2. Steps execute in order, with support for branching, loops, parallel execution, and AI reasoning
  3. Each step can reference data from the trigger and previous steps using data access variables

Triggers

TypeDescription
Colossal triggersFire on Colossal commerce events — orders, products, payments, invoices, customers, notifications
Storefront triggersFire on cart and checkout events
App triggersFire when an external app sends an event to your project
Scheduled triggersFire on a cron schedule or at recurring intervals
See trigger modes to understand how Colossal app trigger timing affects checkout and cart flows.

Step types

StepDescription
App actionExecute an action from any installed app
PromptLLM-powered analysis with access to read and write tools, gated by per-tool approval
ApprovalPause for a human to approve or deny before continuing
ChoiceBranch based on conditions
DelayWait for a duration
ParallelRun branches concurrently
MapLoop over an array
Wait for eventPause until an event arrives from any app
TransformReshape data into a typed output object
EndStop the agent and record the outcome

Apps

Apps provide triggers (events that start agents), actions (operations agents can execute), and tools (the LLM-callable form of those actions). The Agent Builder composes them from your installed apps into agents. The built-in Colossal app ships with every project. It provides:
  • Triggers: 40+ commerce events (ORDER_CREATED, CHECKOUT_COMPLETED, PAYMENT_FAILED, etc.) plus scheduled triggers (cron and interval). See the Colossal and Storefront triggers references.
  • Actions: customers, products, orders, inventory, invoices, email, payments, and HTTP operations, plus cart and checkout. See the Colossal and Storefront actions references.
  • Tools: a curated set of LLM-callable functions for the AI Reasoning step and the chat agent. See the tools reference.

Third-party apps

Install apps from the Colossal dashboard to connect external services. Each app adds its own actions to the Agent Builder, and every action is automatically available as a tool to the LLM inside an AI Reasoning step or the chat agent.
CategoryApps
PaymentStripe, PayPal
MarketingKlaviyo, Mailchimp, Loops
NotificationsSlack
TaxTaxJar
ReviewsYOTPO
More apps are available in the dashboard, with new ones being added regularly. Third-party apps provide actions, tools, and webhook triggers. When you install an app that supports webhooks, its events become available as agent triggers.

Creating agents

Open the Colossal dashboard and describe your automation in the chat:
  • “When an order is confirmed, send a Slack notification to #orders”
  • “Every morning at 9am, check inventory levels and email me a report”
  • “When a customer places an order over $500, flag it as high-value and notify the team”
The assistant generates the agent, validates it, and creates it in your project. You can edit existing agents the same way. Describe what you want to change, or ask it to revert the last change to a previous saved version.

Execution

Each agent run tracks:
  • Status: PENDING, RUNNING, COMPLETED, FAILED, CANCELLED, PAUSED
  • Step-level status: RUNNING, COMPLETED, FAILED, SKIPPED
  • Trigger data: the event or schedule that started the run
  • Step outputs: results from each completed step
Execution is durable. If a step fails, the engine retries automatically. Monitor runs in the dashboard or query them via the GraphQL API.