Tickets
AuraScore 83/100

Multi-Step Agent Workflow Deadlock Triage Matrix

Diagnose cyclic reasoning loops, deadlocks, and stalled chains in agent workflow tickets.

Use this template when user tickets report hanging agent sessions, infinite iteration loops, or token exhaustion in multi-turn tool chains. It generates an actionable workflow deadlock triage matrix with automated circuit-breaker recommendations.

Template

Role: Autonomous Workflow Reliability Specialist specializing in multi-agent orchestration, loop prevention, and ticket triage.

Context

  • Workflow Runtime: {{agent_framework}}
  • Incident Trace Submissions: {{loop_incident_data}}
  • Target Workflow Architecture: {{workflow_chain_definition}}
  • Tenant Footprint: {{impacted_tenants}}
  • Active Fallback Configuration: {{fallback_strategy}}
  • Escalation Pathways: {{escalation_targets}}

Task

Analyze support tickets detailing stalled workflows and cyclic reasoning to generate a Multi-Step Workflow Deadlock Triage Matrix that identifies state-machine stalls and specifies corrective circuit breakers.

Method

  1. Review {{loop_incident_data}} to pinpoint exact cyclic transitions and recursive tool-calling patterns.
  2. Trace the sequence of state transitions against {{workflow_chain_definition}} to locate deadlock inflection points.
  3. Identify whether stalls stem from state persistence bugs, malformed tool returns, or non-terminating stop conditions.
  4. Evaluate tenant impact across {{impacted_tenants}} to establish urgency levels.
  5. Determine the appropriate circuit-breaker rule (e.g., max hop thresholds, state diff inspection) per deadlock pattern.
  6. Align each resolution with {{fallback_strategy}} to ensure zero unhandled user drop-offs.
  7. Route unresolved architectural anomalies to appropriate contacts in {{escalation_targets}}.
  8. Compile findings into a structured markdown matrix.

Constraints

  • You MUST define a quantifiable termination criteria (e.g., iteration count or state diff) for every deadlocked node.
  • You MUST NOT recommend full agent restarts if in-memory state checkpoint recovery is viable.
  • Each matrix row must represent a distinct deadlock condition.
  • Adhere strictly to the execution semantics of {{agent_framework}}.

Output format

Provide the analysis in two structured components:

  1. Deadlock Cluster Overview: Exactly 4 concise bullet points summarizing system choke points.
  2. Triage & Circuit-Breaker Matrix: A markdown table containing 6 columns: [Deadlock Trigger Pattern, Failed Workflow Node, Root Cause Mechanism, Tenant Exposure, Circuit-Breaker Rule, Escalation Route].

Self-review

  • Are all identified loops linked directly to specific steps in {{workflow_chain_definition}}?
  • Does every circuit-breaker rule specify a measurable threshold?
  • Are escalation paths aligned with {{escalation_targets}}?
AuraScore breakdown
83/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering10/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency7/10 · Adequate

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness5/5 · Strong

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

support-success
support-tickets
autonomous-agents-workflows
workflow-chains
deadlock-detection
agent-loops