Create provenance-tracked inbox items from URLs, files, or inline text.
npx skills add https://github.com/Miosa-osa/canopy --skill seed-miosa-osa
Please help me install this Agent Skill. Skill: seed Source: https://github.com/Miosa-osa/canopy/tree/main/library/skills/processing/seed Command: npx skills add https://github.com/Miosa-osa/canopy --skill seed-miosa-osa
Seed sources into your knowledge pipeline by creating provenance-tracked inbox items from URLs, files, or inline text, ensuring no untracked data enters the system.
Seed a new source by passing a URL, file path, or inline text to create a provenance-tracked inbox item.
High-intent search queries and answers about installing and using this skill.
To ingest external data into a provenance-backed knowledge pipeline, you can pass URLs, file paths, or inline text to create structured inbox items. The system automatically handles acquisition, format detection, and metadata extraction for durable downstream storage.
The best way to track provenance metadata for ingested URLs and files is to generate structured seed items during acquisition. This process automatically creates comprehensive provenance records containing the source URL, acquisition time, method, and content hash.
You can deduplicate content during the knowledge ingestion workflow by hashing incoming items and linking them to existing seeds. This prevents duplicate data from entering the inbox and ensures untracked data does not pollute the pipeline.
Yes, the seed ingestion process supports inline text and clipboard inputs alongside URLs and files. It normalizes these varied input formats, generates frontmatter metadata, and writes structured items into a durable inbox for downstream processing.
A key limitation when normalizing external content for a provenance inbox is that the process relies on automated format detection and content hashing. Complex or non-standard input formats may require manual normalization before successful seed creation and deduplication.
kinnerkarmanishConvert raw inputs into structured inbox items with provenance metadata.
misha-jattTriage Markdown files in /Needs_Action, classify priority, and move completed items to /Done.
indranilbanerjeePersist brand knowledge to a vector store for semantic retrieval across sessions.
gaiastudio-aiFetch external documents and write provenance-stamped markdown into the Brain index.
