Skip to main content

Testing Sandbox

The Testing Sandbox lets you simulate rule execution before affecting real members. See exactly what would happen—safely.

Why Test First?​

Rules can affect hundreds of members instantly. A mistake might:

  • Add wrong roles to everyone
  • Remove important roles accidentally
  • Trigger unexpected cascades

The sandbox eliminates this risk.

How to Test​

Step 1: Select Test Roles​

In the sandbox, select roles to simulate a member having.

Example: To test "Verified + Level 5":

  1. Click "Verified" to select
  2. Click "Level 5" to select

Step 2: Run Test​

Click "Run Test" to simulate.

Step 3: Review Results​

The sandbox shows:

  • Roles added: What would be given
  • Roles removed: What would be taken
  • Rules triggered: Which rules fired and in what order

Step 4: Iterate​

Adjust test roles and run again for different scenarios.

What to Test​

Common Scenarios​

  • New member (no roles except @everyone)
  • Verified member with basic roles
  • VIP member with premium roles
  • Staff member with mod roles

Edge Cases​

  • Member with all roles
  • Member with conflicting roles
  • Member with only @everyone

High-Population Roles​

Test with roles many members have before going live. A mistake here affects more people.

Understanding Results​

"Roles Added"​

Roles that would be assigned.

"Roles Removed"​

Roles that would be taken away.

"Rules Triggered"​

Shows the cascade:

  1. Rule A triggered first (priority 0)
  2. Rule B triggered second (priority 1)
  3. etc.

No Changes​

If no rules matched, either:

  • Conditions don't match test roles
  • Rule is working correctly by not matching

Loop Detected​

If the result shows a red These role conditions loop notice, the selected roles make your rules undo each other without end. The real bot behaves the same way: it stops evaluating and changes nothing for that member. Fix the conflict before going live.

Testing Cascades​

The sandbox accurately simulates cascading:

  1. Initial roles selected
  2. Rule A fires → adds Role X
  3. Sandbox checks if Role X triggers Rule B
  4. Rule B fires → adds Role Y
  5. Final result shows all changes

Tips​

  • Always include @everyone — Every member has it
  • Test after changes — Old results are invalid after modifying rules
  • Test incrementally — Build complex rules step by step, testing each
  • Document expected results — Compare with actual to catch surprises

Limitations​

  • Doesn't test permissions — Role hierarchy issues won't show
  • Instant results — Real execution has slight latency
  • Current rules only — Save changes before testing new versions. For a member-count estimate of an unsaved draft, use Check impact in the rule editor instead.
Last updated on