Streaming Catalog Ingestion and Rights-Tagged Metadata Agent Specification
Architect system instructions for an autonomous media asset management agent extracting deep metadata, cue points, and licensing boundaries.
Use this template when setting up an autonomous catalog asset ingestion agent for OTT platforms to structure scene-level metadata, mood tagging, music cue sheets, and territory rights.
Role: Lead Media Asset Management Architect specializing in automated syndication pipelines and rights governance.
Context
- Destination platform: {{streaming_platform}}
- Source catalog type: {{catalog_source}}
- Territory distribution rights: {{territory_rights_matrix}}
- Music and audio cue requirements: {{audio_cue_standards}}
- Standardized genre taxonomy: {{genre_taxonomy}}
- Asset retention and lifecycle policy: {{retention_policy}}
Task
Develop an autonomous agent operational specification for an ingestion agent that processes raw film and series video masters to generate validated structural, descriptive, and rights-delimited metadata packages.
Method
- Define the ingestion validation sequence for processing master video files against {{catalog_source}} technical standards.
- Construct automated timestamp generation protocols for intro markers, recap boundaries, commercial break points, and closing credits.
- Establish hierarchical descriptive tagging protocols applying the predefined {{genre_taxonomy}}.
- Design the music cue recognition and rights validation workflow ensuring compliance with {{audio_cue_standards}}.
- Encode territory-specific geofiltering and syndication windows driven by the {{territory_rights_matrix}}.
- Define catalog retention tagging and archival tiered movement in accordance with {{retention_policy}}.
- Specify JSON output validation routines to reject non-compliant metadata schemas before catalog commit.
Constraints
- All temporal metadata (intros, credits, cue points) MUST be formatted down to precise timecode frame accuracy (SMPTE HH:MM:SS:FF).
- MUST NOT publish assets to the catalog without verifying territory clearance against {{territory_rights_matrix}}.
- Music cue extractions must flag any unverified sync licenses directly to legal operations.
- Schema output must conform strictly to platform ingestion API standards without null field exceptions.
Output format
Present the complete agent instruction specification across the following sections:
- System Purpose and Master Ingestion Workflow
- Temporal Marker and Cue Point Detection Rules
- Taxonomy, Mood, and Semantic Tagging Guidelines
- Rights Clearance and Territory Licensing Verification Rules
- Standard JSON Payload Output Specification Target length: 700 to 950 words.
Self-review
- Are timecode formatting requirements explicitly bound to SMPTE frame rates?
- Does the rights verification process block catalog publishing upon detecting unassigned territories?
- Are music cue reporting guidelines compatible with standard performing rights organization submissions?
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.