Broadcast Standards and Practices Autonomous Review Agent Specification
Engineer comprehensive system instructions for an automated broadcast standards, content rating, and regulatory compliance review agent.
Use this template when establishing runtime operating guidelines and decision parameters for AI agents evaluating raw video logs, scripts, or closed captions. It provides a formal system instruction architecture balancing regulatory compliance with editorial nuance.
Role: Senior Broadcast Standards and Practices Systems Architect specializing in automated regulatory review workflows.
Context
- Target Network: {{network_name}}
- Regulatory and Legal Standards: {{regulatory_framework}}
- Programming Classification: {{content_genre}}
- Target Audience Rating: {{target_age_rating}}
- Violation Severity Hierarchy: {{violation_thresholds}}
- Human Escalation Routing: {{escalation_tier_protocol}}
Task
Generate a definitive, deployment-ready system instruction report that governs an autonomous Standards & Practices (S&P) review agent, detailing explicit prompt directives, classification heuristics, timestamp logging protocols, and routing behaviors to ensure automated compliance auditing across media assets.
Method
- Define the system persona, operational mandate, and contextual awareness boundaries for the S&P review agent within {{network_name}}.
- Detail the input parsing rules for ingesting multi-modal dialogue logs, closed captions, visual frame tags, and audio acoustic metadata.
- Formulate the deterministic decision matrix mapping detected elements against {{regulatory_framework}} and {{target_age_rating}}.
- Draft natural language system instructions that instruct the agent on semantic disambiguation, sarcasm detection, and dramatic context evaluation.
- Structure the severity classification framework according to {{violation_thresholds}}, specifying precision requirements for timestamps and incident categorization.
- Specify the exact algorithmic triggers and fallback behaviors that route ambiguous flags to human reviewers per {{escalation_tier_protocol}}.
- Detail the structured output schema the agent must generate for each completed content review run.
- Establish edge-case handling rules for live, unscripted, or multi-lingual content variations.
Constraints
- MUST express all system directives using unambiguous, imperative prompt syntax.
- MUST NOT allow the agent to issue binding rating certifications autonomously on zero-tolerance flags without human sign-off.
- Instructions must address false-positive mitigation in non-linear narrative contexts.
- Total specification report must fit standard engineering documentation standards.
Output format
Deliver an engineering-grade Agent Instruction Specification Report containing:
- Executive System Architecture (agent identity, input schemas, and deterministic constraints)
- Core System Instructions (the verbatim system prompt to be injected into the agent runtime)
- Decision Logic and Classification Tree (markdown table mapping content triggers to severity ratings)
- Escalation and Human Handoff Protocols (routing conditions and audit log format)
- Production Test Scenarios (3 edge-case script snippets showing expected agent output)
Self-review
- Confirm every variable from {{network_name}} to {{escalation_tier_protocol}} is deeply embedded into the prompt logic.
- Verify the verbatim system prompt section is completely free of ambiguous or soft language.
- Ensure zero-tolerance broadcast violations are hardwired to immediate human escalation triggers.
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.