Platform Features

Beyond the core workflow engine and nodes, Agents includes advanced features for building sophisticated, production-ready automations.

Feature Categories

Temporal Controls

Wait Node Deep Dive Delay workflow execution for precise timing control. Support for durations from seconds to years.

Scheduling & Timers Create time-based workflow triggers using cron expressions with timezone support.

Reliability & Observability

Dead Letter Queue Handle and recover from message delivery failures with automatic failure isolation.

Monitoring & Logs Track execution status, view detailed logs, and monitor workflow health.

Workflow Management

Versioning Manage workflow versions with controlled releases, rollbacks, and revision tracking.

Feature Overview

Production Considerations

Reliability Features

The platform includes several features designed for production reliability:

1. Transactional Outbox - Messages are persisted before sending, preventing data loss
2. Retry Policies - Configurable retry strategies with exponential backoff
3. Dead Letter Queues - Failed messages are preserved for investigation
4. Execution Tracking - Full visibility into workflow execution state

Scalability Features

Built to handle varying workloads:

1. Event-Driven Architecture - Scales based on event volume
2. Batch Processing - Efficient handling of multiple items
3. Parallel Execution - Independent branches execute concurrently
4. Rate Limiting - Use Wait nodes to control API call rates

Observability Features

Tools for understanding system behavior:

1. Execution Logs - Detailed logs for each workflow run
2. Node-Level Tracking - Per-node execution status and outputs
3. Error Reporting - Clear error messages with context
4. Timeline Views - Visual execution history

Getting Started with Features

For New Users

Start with these fundamental features:

1. Monitoring - Understand how to view execution status
2. Error Handling - Learn about retry policies
3. Versioning - Know how to manage workflow changes

For Production Releases

Focus on reliability features:

1. DLQ Management - Set up monitoring for failed messages
2. Scheduling - Configure time-based workflows correctly
3. Wait Node - Use for rate limiting and timing control

Feature Interactions

Wait + Scheduling

Combine scheduled triggers with wait nodes:

[Scheduled: Daily 9 AM] → [Load Data] → [Wait: 1 hour] → [Send Report]

Monitoring + DLQ

Use monitoring to detect DLQ issues:

1. Monitor execution failure rates
2. Review DLQ for patterns
3. Address root causes
4. Retry failed messages

Versioning + Monitoring

Safe release workflow:

1. Create new version
2. Release to limited scope
3. Monitor execution success rates
4. Roll out fully or rollback

Best Practices

1. Plan for Failures

• Configure appropriate retry policies
• Design error paths in workflows
• Monitor DLQ regularly

2. Use Appropriate Timeouts

• Standard nodes: 30 seconds
• AI nodes: 120 seconds
• Adjust based on expected execution time

3. Version Thoughtfully

• Use meaningful release notes
• Test before releasing
• Keep rollback path available

4. Monitor Actively

• Review execution logs regularly
• Set up alerts for failure patterns
• Track performance metrics over time

Related

• Node Reference
• Guides
• Troubleshooting