Workflow chains
AuraScore 81/100

Institutional Research Ethics Protocol Screening Workflow

Structure an autonomous agent triaging, risk-scoring, and compliance-routing pipeline for institutional review board submissions.

Use this template to design an automated pre-screening workflow for institutional research applications, ensuring systematic risk classification, policy compliance checks, and regulatory routing.

Template

Role: Senior Research Compliance Technologist specializing in automated institutional oversight workflows.

Context

  • Research Institution Profile: {{institution_type}}
  • Regulatory and Funding Mandates: {{grant_funding_bodies}}
  • Institutional Risk Taxonomy: {{risk_classification_tiers}}
  • Audit and Traceability Standard: {{audit_logging_framework}}
  • Human Escalation Thresholds: {{escalation_thresholds}}

Task

Design an autonomous compliance routing and pre-screening workflow chain that systematically evaluates research proposals, extracts human-subject risk factors according to {{risk_classification_tiers}}, validates compliance against {{grant_funding_bodies}}, and prepares structured advisory dossiers for review boards.

Method

  1. Structure the intake parser agent to extract methodology, participant demographics, data privacy plans, and consent protocols from incoming research proposals.
  2. Design the regulatory evaluation sub-chain to test proposal parameters against {{grant_funding_bodies}} mandates.
  3. Implement the risk scoring agent that maps extracted methodologies to {{risk_classification_tiers}}.
  4. Formulate the triage routing logic that determines whether a protocol qualifies for exempt, expedited, or full-board review.
  5. Build the audit trail generator agent in alignment with {{audit_logging_framework}} to record decision paths.
  6. Embed the automated escalation interrupt based on {{escalation_thresholds}} for immediate human handover.
  7. Construct the final dossier generation agent to synthesize an itemized pre-review report for the ethics committee.

Constraints

  • Agent classifications MUST NOT make final legal or ethical determinations without logging rule references.
  • The pipeline MUST trigger an immediate hard stop and human escalation whenever {{escalation_thresholds}} are breached.
  • Sensitive researcher and participant data MUST be handled in strict accordance with {{institution_type}} governance policies.
  • Audit logs must be immutable and deterministic.

Output format

Provide the workflow framework organized into four structured components:

  1. Multi-Stage Pipeline Graph (intake, classification, validation, routing, and output nodes)
  2. Decision Logic & Rule Matrices (deterministic criteria for risk scoring and tier mapping)
  3. Escalation & Intervention Protocols (precise parameters executing human handoffs)
  4. Review Dossier Schema (template layout for ethics committee summary packages)

Self-review

  • Confirm that every compliance rule directly maps to a mandate defined in {{grant_funding_bodies}}.
  • Verify that edge-case ambiguity automatically triggers the {{escalation_thresholds}} protocol.
  • Ensure the output audit schema complies completely with {{audit_logging_framework}}.
AuraScore breakdown
81/100Provisional
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 engineering12/12 · Strong

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.

ai-agents
agents-workflows
education-research
workflow-chains
research-ethics
irb-compliance