Operational Rules are high-priority directives that govern how agents execute tasks across your organization. These rules override specialty guidelines and apply to all agents in the same organization, ensuring consistent operational procedures. Example: For agents managing disk space, you might require them to exhaust cleanup options (clear /tmp, rotate logs, remove old backups) before provisioning additional capacity.

​
Managing Operational Rules

Go to Accounts → Operational Rules and click Create Rule to add a new entry. Use the action buttons on each row to edit or delete rules. Operational Rules

​
Name

A clear identifier for the rule. Example: Restart services after config changes

​
Organization

By default, rules apply to all organizations. Select a specific organization to restrict the rule to only that org’s agents.

​
Description

The actual rule that agents must follow. Operational Rules take precedence over specialty instructions. Write rules that are concise, specific, and focused on a single operational concern. ⚠️ Warning: Overly strict or conflicting rules may prevent agents from completing tasks. Test new rules carefully before rolling them out. Example: Always restart processes after updating their configuration files With this rule, an agent modifying an nginx config will automatically run sudo systemctl restart nginx after making changes—regardless of the agent’s specialty.

​
Common Use Cases

Service Management
Restart services after configuration changes. Validate config syntax before applying. Create backups before modifying critical files. Resource Management
Attempt cleanup before provisioning additional resources. Check utilization thresholds before scaling. Document resource changes in tracking systems. Security & Compliance
Require approval for production deployments. Enforce logging for privileged operations. Validate credential rotation procedures. Change Management
Create snapshots before infrastructure modifications. Notify teams before impactful changes. Follow rollback procedures on failure.

​
Best Practices

Start with broader guidelines and refine based on agent behavior. Test rules in isolation to identify conflicts. Keep rules technology-agnostic when possible—focus on outcomes rather than specific commands. Review rule effectiveness regularly and deprecate what you don’t use. Document why each rule exists for future reference. Use Blacklist to prevent specific commands, and Operational Rules to guide behavior. Combine with Specialties for domain-specific workflows that don’t need organization-wide enforcement.