Authoring & Executing
The platform keeps designing rule sets separate from running them. This page explains what belongs to each side and how a design you author turns into a workload that runs.
The authoring side
Authoring is everything you do to design and refine your logic. You work inside a project, building rule sets made of rules, each combining a condition with one or more actions. You can reuse conditions and action sets, pull in extensions and templates, and let the AI Template Generator draft logic from a prompt. Nothing here runs against real input yet — you are describing what should happen.
The executing side
Executing is where a rule set actually runs, inside an environment. An environment is an isolated runtime that belongs to your organization and carries its own settings, variables, schedules, storage, and logs. When you run a rule set, you provide input, each rule's condition is evaluated, and the actions of every matching rule run. Runs can start on demand, on a schedule, or through the API, and each run is recorded in Logs & Run History.
Why the two are separate
Separating design from execution lets one rule set run in many environments without being rebuilt — for example a test environment and a production one, or separate environments per customer. It also explains where things live:
| Belongs to authoring | Belongs to an environment (executing) |
|---|---|
| Projects, rule sets, rules, conditions, actions | Run input and the resulting context |
| Reusable conditions and action sets | Environment variables and settings |
| Extensions and templates you install | Schedules, stored files, and run logs |
| Members who can edit the project | Roles that govern who can run or manage the environment |
From a design to a run
- You finish a rule set in a project.
- You select an environment to run it in and supply the input it expects.
- The rule set evaluates each rule's condition against that input.
- Matching rules run their actions, which may call services, transform data, or pause for review.
- Results, files, and logs are recorded against the run's correlation id for you to review.