Literature review
AuraScore 79/100

Multi-Agent Workflow Orchestration and State Graph Literature Synthesis Plan

Plan a comprehensive literature review assessing state graph architectures, consensus mechanisms, and multi-agent coordination chains.

Use this prompt to outline a research plan for investigating academic papers and technical specs on workflow DAG orchestration, state persistence, and inter-agent communication protocols. It is designed for researchers and staff engineers standardizing multi-agent workflows.

Template

Role: Staff Research Scientist in Multi-Agent Systems and Distributed Workflow Automation.

Context

  • Orchestration Paradigm: {{orchestration_paradigm}}
  • Consensus and Handoff Mechanisms: {{consensus_mechanisms}}
  • State Persistence Infrastructure: {{state_persistence_engine}}
  • Target Benchmark Corpus: {{benchmark_corpus}}
  • Target Academic Databases: {{academic_databases}}
  • Review Sprint Duration: {{review_sprint_duration}}

Task

Design a rigorous, publication-ready literature review plan that analyzes, synthesizes, and maps existing academic literature on multi-agent execution graphs, agent-to-agent message passing, and deterministic state transitions under {{orchestration_paradigm}}.

Method

  1. Formulate exact search strings and Boolean filters targeting {{academic_databases}} focused on graph-based agent orchestration.
  2. Establish systematic criteria for isolating peer-reviewed literature on {{consensus_mechanisms}} and dynamic task delegation.
  3. Design a data extraction schema covering state serialization efficiency, token cost per task, deadlock occurrence, and scalability limits.
  4. Structure a deep-dive evaluation process to assess how papers address memory leakage and state synchronization in {{state_persistence_engine}}.
  5. Categorize reviewed workflows into centralized hierarchical, decentralized gossip, and directed acyclic graph (DAG) execution topographies.
  6. Cross-reference empirical metrics in selected papers against standard multi-agent benchmarks in {{benchmark_corpus}}.
  7. Synthesize trade-offs between deterministic workflow engines and open-ended autonomous agent decision loops.
  8. Build the phased delivery milestones, task owners, and artifact checklist across {{review_sprint_duration}}.

Constraints

  • Focus exclusively on multi-agent workflow chains, state coordination, and communication protocols.
  • MUST define explicit quantitative thresholds for paper inclusion regarding reproducibility on {{benchmark_corpus}}.
  • MUST NOT prioritize single-agent chain-of-thought literature unless it directly impacts multi-agent state handoffs.
  • Every extraction step must account for state consistency trade-offs.

Output format

  • Literature Review Scope Statement: 200 words defining target multi-agent paradigms and research questions.
  • Search and Extraction Protocol: Detailed database query rules, inclusion/exclusion matrix, and screening stages.
  • Review Timeline and Sprint Plan: Gantt-style structured text plan spanning {{review_sprint_duration}} with resource allocation.
  • Synthesis Framework and Deliverable Structure: Outlined chapters for the final synthesis report with page budget estimates.

Self-review

  • Does the method clearly address state persistence across distributed agent execution?
  • Are the search protocols tailored specifically to {{academic_databases}}?
  • Does the plan enforce empirical verification using {{benchmark_corpus}}?
AuraScore breakdown
79/100Provisional
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 efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness3/5 · Adequate

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

research-analysis
research-literature
autonomous-agents-workflows
multi-agent-systems
workflow-orchestration
literature-review