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