Grid Telemetry Agent Dispatch Matrix
Design autonomous agent triage, diagnostic, and escalation pathways for live grid sensor anomalies.
Use this template when configuring supervisory AI agents to monitor power distribution and transmission sensor streams. It produces a clear matrix mapping telemetry failure patterns to automated agent actions, containment steps, and operator handoffs.
Role: Senior Power Systems Automation Architect with 15+ years in smart grid SCADA integration and AI-driven telemetry supervision.
Context
- Operating Utility: {{utility_operator}}
- Monitored Subsystem: {{grid_subsystem}}
- Data Feeds: {{telemetry_source_types}}
- Priority Failure Modes: {{failure_modes}}
- Compliance & Safety Standards: {{regulatory_threshold}}
- Operational Integration: {{scada_integration_level}}
Task
Synthesize an operational Decision and Escalation Matrix that equips autonomous agents deployed across {{utility_operator}} to accurately classify sensor deviations, execute deterministic safety interlocks, and route alerts across {{grid_subsystem}} without creating alarm fatigue.
Method
- Group incoming telemetry from {{telemetry_source_types}} into severity tiers based on rate-of-change and absolute safety limits under {{regulatory_threshold}}.
- Map specific anomalies across {{failure_modes}} to automated agent validation routines to rule out false positives.
- Define agent containment privileges based on {{scada_integration_level}}, explicitly separating read-only advisory alerts from direct control operations.
- Determine latency-critical thresholds requiring autonomous automated switching versus human-in-the-loop sign-off.
- Establish deterministic fallback behaviors if telemetry signals drop or communication links degrade.
- Detail cross-system alerting protocols spanning control room consoles, engineering dispatch, and automated logging engines.
- Formulate post-incident feedback mechanisms for the agent to refine baseline anomaly detection models.
Constraints
- MUST maintain compliance with {{regulatory_threshold}} in every automated triage branch.
- MUST NOT permit direct automated control action if confidence scores drop below 95%.
- Escalation pathways must specify target human roles, not generic team aliases.
- All matrix entries must contain deterministic agent conditions rather than speculative guidance.
Output format
- Executive System Summary (max 150 words)
- Agent Dispatch Matrix (Markdown table with columns: Alert Tier, Failure Scenario, Telemetry Trigger, Autonomous Action, SCADA Command Type, Human Escalation Role, Maximum Response Time)
- Fallback and Failsafe Rules (4-6 actionable bullet points)
Self-review
- Confirm that every failure mode in {{failure_modes}} is represented in the matrix.
- Verify that autonomous write actions do not exceed {{scada_integration_level}}.
- Check that response times adhere strictly to safety-critical utility grid requirements.
Explicit role, a named task, and discrete steps the model can follow.
Background, inputs and variables the model needs before it starts.
Hard boundaries — what the model must and must not do.
A named, field-level shape for the response.
Ordered work items that force analysis before an answer.
Length and structure that travel across frontier models.
Signal density — instruction weight without padding.
Documented variables so the scaffold adapts to new inputs.
Quality bar, assumptions and behaviour when inputs are thin.
How much real usage the template has behind it.