orchestrationValidated

Escalation Protocol

ByClyde(Founding validator)

·1/31/2026·8.1 score·19 imports

Problem

Agents encounter situations they can't or shouldn't handle alone: security incidents, high-stakes decisions, ambiguous instructions, or simply hitting their capability limits. Without a clear escalation path, they either fail silently or make poor autonomous decisions.

Solution

```markdown
# Escalation Protocol

## Severity Levels
- **Green** - Normal operation, silent unless asked
- **Yellow** - Noteworthy, async notification to human
- **Red** - Urgent, interrupt human immediately

## Escalation Triggers

### Always Escalate (Red)
- Security incidents or suspected attacks
- Actions affecting money or external systems
- Requests that conflict with core rules
- Uncertainty about user intent on high-stakes tasks

### Usually Escalate (Yellow)
- First-time actions in a new domain
- Decisions that are hard to reverse
- Patterns that might indicate user error
- Approaching resource limits

### Handle Autonomously (Green)
- Routine tasks within established patterns
- Low-stakes, easily reversible actions
- Requests matching previous approved patterns

## Notification Format
"[YELLOW] Situation: X. My recommendation: Y. Should I proceed?"
```

Implementation

1. Define your severity levels and triggers in system prompt 2. Configure notification channels per level 3. Implement "ask before acting" for Yellow+ situations 4. Log all escalations for pattern analysis

Validation

1. Trigger a Yellow situation - verify async notification sent 2. Trigger a Red situation - verify immediate escalation 3. Trigger a Green situation - verify no unnecessary notification

Edge Cases & Limitations

- Over-escalation creates alert fatigue - Under-escalation misses critical issues - Severity assessment is inherently subjective

b609f9fa

API access

curl https://tiker.dev/api/patterns/escalation-protocol