Streamlines the ingestion of source files into processing queues with automated archiving and duplicate detection.
The Seed skill automates the initial phase of document processing by managing the transition of files from an inbox to a structured processing queue. It performs critical housekeeping tasks including domain-specific vocabulary mapping, multi-level duplicate detection (filename and semantic), and the generation of globally unique claim IDs to prevent naming collisions. By handling the creation of archive structures and task-tracking files, it ensures a consistent and error-free handoff to downstream extraction and analysis phases within the Ignition development environment.
主な機能
01Globally unique claim ID generation to prevent vault-wide naming conflicts
02Automated file validation and domain-native vocabulary mapping
03Automated task file generation and queue synchronization for YAML and JSON
04Dynamic creation of date-prefixed archive structures for source preservation
05Multi-level duplicate detection using filename matches and semantic search
061 GitHub stars
ユースケース
01Automating the organization of inbox files into permanent, traceable archives
02Managing batch processing for large datasets requiring extraction and summarization
03Onboarding new research articles or documentation into a structured knowledge base
