AIgency scheduler
The scheduler repeatedly finds structurally valid, dependency-satisfied tasks and starts them within concurrency and capacity bounds.
AgentDock Pro
AIgency is available only in AgentDock Pro.
aigency-service.ts implements the scheduler as a repeated pass rather than a persistent loop: whenever a task settles (completes, fails, or otherwise changes state), the very next scheduling pass re-evaluates what's now eligible to run.
The gates a task must clear
- Dependency validation — every prerequisite must be genuinely
completed, not merely no-longer-blocking. - Pause gate — nothing new starts while the run is paused.
- Concurrency gate — both the run's global concurrency ceiling and any per-agent ceiling must have room.
- Capacity gate — a task whose target agent is mid-capacity-wait isn't started.
Verified bounds
| Bound | Value |
|---|---|
| Default concurrency | 2 |
| Maximum configured concurrency | 6 |
| Maximum retries per task | 5 |
| Maximum fallback agents per task | 2 |
| Interaction expiry | 24 hours |
| Mid-run silence stall warning | 15 minutes |
These are hard caps enforced in the scheduler itself, not just UI-suggested defaults — configuring a higher value than the cap clamps to it rather than erroring.