Getting Started
A practical, end-to-end walkthrough: sign in, pick an organization, create a project, author a rule set with one rule, choose an environment, run it, and read the logs.
The AI Rule Engine separates authoring (designing rule sets) from executing (running them). You build rule sets in the app and run them in an environment. If the terms below are unfamiliar, skim Core Concepts first, then come back here. For a bird's-eye view, see How the Platform Works.
1. Sign in
Open the app and sign in with one of the supported providers. Signing in establishes who you are and which organizations and projects you can access. See Authentication for the sign-in options.
2. Create or choose an organization
An organization is the billing and ownership boundary that contains your projects, environments, members, plan, and credits. After signing in, select an existing organization from the organization switcher, or create a new one. The organization you choose determines which plan limits and credit balance apply to your work. Learn more under Organizations and Plans & Tiers.
3. Create a project
A project groups related rule sets, conditions, and action sets. Create a project, give it a name and an optional description, and leave the project type as Standard (the other types, Extension and Template, support marketplace authoring). The project owner and members control who can edit its contents. See Projects for details.
4. Author a rule set with a rule
Inside the project, create a rule set. A rule set is an ordered collection of rules plus a context definition that declares the inputs the rules expect. You can control how the rules run: evaluate them in order or in parallel, optionally randomize their order, and optionally cap how many rules may match. Full reference: Rule Sets.
Add a single rule, which has two parts:
- A condition that decides whether the rule matches. A condition is a group of clauses combined with All, Any, or None logic; each clause compares a context value (referenced by a context key) against a value using an operator such as Equals or Contains. See Conditions.
- One or more actions that run when the rule matches. Actions can call an API, transform data, write to the context, log, run an AI prompt, and more. Browse the catalog under Actions Overview and AI Actions.
A rule references either a saved condition or an inline one, and either saved action sets or inline actions — one or the other, not both. For a visual way to assemble all of this, try the Workflow Builder.
5. Choose or create an environment
Rule sets are authored once but run inside an environment. An environment is an isolated place to run rule sets that belongs to your organization and holds its own settings, variables, schedules, and logs. Create one (or select an existing one) so it can accept run requests. Environments run in one of three hosting options — Dedicated, Shared, or Free tier — described under Hosting Options and Environments.
6. Run the rule set
With an environment selected, run the rule set. You supply input that satisfies the context definition; each rule's condition is evaluated against that context and the actions of every matching rule run, honoring the ordering you configured. You can run on demand from the app, on a schedule, or with the API. The mechanics are covered in Running Rule Sets.
7. View the logs
After a run, open the logs to see which rules matched, what each action did, and any errors. Logs can be viewed by recent activity or grouped by correlation id. See Logs & Run History.