Rule Engine overview

The Rule Engine evaluates declarative rules against any JSON payload. A rule has a condition and a then/else block. Rules are organized into rule sets, persisted, versioned, and executed via REST. It powers pricing decisions, eligibility checks, dynamic scoring, fraud gates, and any custom logic that needs to change without redeploying code.

What it covers

• Pricing rules — Bulk discounts, channel-specific markups, dynamic margins based on competitor data.
• Eligibility checks — Whether an order can ship to an address, whether a customer qualifies for a promo, whether a lead matches an ICP.
• Scoring — Lead scoring weights, fraud risk, content recommendations.
• Custom business logic — Anything you'd otherwise write in a one-off service. Rule authors get YAML or JSON; engineers don't have to redeploy.

Surface area

Two controllers expose the engine:

• /rule-engine/* — Stateless execution and validation.
• /rules/* — Rule sets (CRUD, versions, clone, import/export), helpers (custom functions), AI generation, analytics.
• /rule-sets/* — Lower-level rule-set CRUD.

Rule authoring: write YAML or JSON, validate, save as a rule set, execute by ID.

Rule shape, in brief

version: 2.0
rules:
  - id: BULK_DISCOUNT
    when: order.lineCount >= 10
    then:
      apply_discount:
        type: percent
        value: 5
    else:
      apply_discount:
        type: percent
        value: 0

The same rule in JSON:

{
  "id": "BULK_DISCOUNT",
  "when": { ">=": ["order.lineCount", 10] },
  "then": { "apply_discount": { "type": "percent", "value": 5 } },
  "else": { "apply_discount": { "type": "percent", "value": 0 } }
}

Rules can iterate (for), switch (switch/cases), set variables (vars), and call helper functions (COUNT, FILTER, SUM, plus org-defined helpers).

Where each piece lives

Topic	Page
Data shape, operators, helpers	Syntax
Natural-language rule generation	AI generation
Validate and dry-run	Testing
Caching and indexed evaluation	Performance

Cross-module use

The engine is used internally by:

• Sales Channel — per-channel pricing rules.
• CRM — audience segmentation, lead routing, scoring weights.
• Storefront — discount eligibility, dynamic pricing.
• Automation — when a flow needs richer evaluation than check_field_value can express, an action calls the rule engine and branches on the outcome.