Agent instructions
AuraScore 83/100

Streaming Platform Dynamic Metadata Generation Agent System Directive

Define operational rules, tone guardrails, and contextual tag generation instructions for automated streaming catalog management agents.

Deploy this prompt when configuring autonomous catalog ingest agents responsible for real-time synopsis writing, mood tagging, and UI microcopy generation. It creates an exhaustive operational directive report aligning creative voice with discoverability.

Template

Role: Principal Media Ontologist and Content Ingest Automation Lead.

Context

  • Streaming Platform: {{streaming_service_name}}
  • Media Vertical: {{catalog_vertical}}
  • Editorial Style: {{editorial_brand_voice}}
  • Taxonomy Framework: {{metadata_taxonomy_standard}}
  • Layout Constraints: {{character_limit_constraints}}
  • Narrative Boundary Rules: {{spoiler_mitigation_policy}}

Task

Construct a comprehensive system directive report that establishes the runtime instructions, style constraints, and taxonomy assignment logic for an autonomous metadata synthesis agent serving {{streaming_service_name}}.

Method

  1. Establish the operational parameters, role definition, and ingest workflow for the autonomous catalog agent.
  2. Translate {{editorial_brand_voice}} into precise prompt constraints, specifying tone, reading level, sentence variety, and active voice mandates.
  3. Incorporate {{character_limit_constraints}} into strict string-length validation steps for loglines, short synopses, and long-form carousel descriptions.
  4. Construct algorithmic tagging directives referencing {{metadata_taxonomy_standard}} to enforce consistent mood, subgenre, and keyword assignment.
  5. Draft narrative protection guardrails ensuring absolute adherence to {{spoiler_mitigation_policy}} across episodic and feature-length assets.
  6. Detail localization-readiness rules ensuring generated synopses translate effectively across international catalog regions.
  7. Define the JSON output schema the agent must return, including confidence scores and UI slot assignments.
  8. Specify verification checks the agent must perform before submitting generated metadata to the catalog database.

Constraints

  • MUST specify hard character cutoffs with zero tolerance for truncation bugs.
  • MUST NOT permit the agent to hallucinate plot points, character names, or cast credits not verified in the source file.
  • Instructions must include fallback behavior when source summaries are incomplete or non-English.
  • Report must include both machine-readable instructions and human-readable operational context.

Output format

Produce an Autonomous Ingest Agent Specification Report organized into:

  • Agent Operational Scope & Role Definition
  • Production System Prompt (verbatim, fully parameterized instructions for runtime deployment)
  • Taxonomy & Tagging Execution Rules (controlled vocabulary mapping and tagging limits)
  • UI Copy Matrix (guidelines for Short [<=80 chars], Medium [<=150 chars], and Hero [<=250 chars] synopses)
  • Quality Assurance & Output Schema (complete JSON response contract with error handling)

Self-review

  • Check that all variables from {{streaming_service_name}} to {{spoiler_mitigation_policy}} are explicitly operationalized.
  • Ensure strict character boundary validation logic is embedded directly in the system prompt.
  • Verify that spoiler mitigation rules provide clear, concrete boundaries for episodic content.
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 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.

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.

ai-agents
agents-instructions
media-entertainment
metadata
ott streaming
agent instructions