How does it prevent duplicate work?

It performs both filename matching and keyword/semantic search to identify if the content has already been ingested or processed, asking for confirmation before proceeding.

Can I use it for files outside the inbox?

Yes. If a file is outside the designated inbox (like a living config file), Seed will create the task and archive folder but leave the original file in its current location.

What is a 'Claim Number' in this context?

It is a globally unique 3-digit ID (e.g., 001, 002) assigned to parts of the source to ensure every generated file has a unique name across your entire project, preventing wiki-link collisions.

What does the Seed skill actually do?

It moves a source file from an inbox to an archive, checks if it has been processed before, and creates a formal task entry in your processing queue to trigger downstream extraction.

Which queue formats are supported?

The skill supports both YAML (ops/queue.yaml) and JSON (ops/queue/queue.json) formats for managing the processing backlog.

Source Seeding & Queue Management

Name: Source Seeding & Queue Management
Author: LopeWale

byLopeWale

0•

生产力与工作流

Automates the ingestion of source files into a processing queue with duplicate detection and structured archiving.

The Seed skill streamlines the initial phase of data processing by managing the transition of source files from an inbox to a formal processing queue. It performs critical housekeeping tasks including semantic duplicate detection, global claim ID numbering to prevent filename collisions, and the creation of standardized task files. By automating the 'claiming' process and moving files to permanent archives, it ensures that research articles, documentation, and transcripts are handled with consistent metadata and a clear audit trail, preventing redundant work in multi-step extraction pipelines.

主要功能

01Structured archiving with date-prefixed folders for permanent source storage

020 GitHub stars

03Globally unique claim ID numbering to prevent vault-wide naming collisions

04Standardized task file generation (YAML/JSON) for downstream automation

05Automated duplicate detection using filename and keyword/semantic search

06Domain-aware vocabulary mapping via derivation manifests

使用场景

01Ingesting research papers or articles into a knowledge management vault

02Automating the hand-off between raw data collection and AI-driven extraction tasks

03Managing a high-volume processing pipeline for technical documentation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add lopewale/ampllabs seed

For use in Claude.ai and ChatGPT