document-storage

Persists structured document records to gbrain with YAML frontmatter and path conventions.

1|Updated Jun 21, 2026
One-click install
npx skills add https://github.com/tapway/shogun-os --skill document-storage-tapway
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: document-storage
Source: https://github.com/tapway/shogun-os/tree/main/skills/general/document-storage
Command: npx skills add https://github.com/tapway/shogun-os --skill document-storage-tapway

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After a document has been interpreted, its structured data needs a consistent, traceable home. This Skill writes document records to gbrain at standardized paths with proper YAML frontmatter, so invoices, quotations, and other files stay organized and retrievable later. ## Core Features & Use Cases - Standardized Storage Paths: Writes records to references/documents/<type>/ directories covering invoices, quotations, legal documents, purchase orders, delivery orders, and other types. - Consistent Frontmatter: Generates pages with title, type, tags, source, stored date, and document_type fields so retrieval works reliably. - Timeline Integration: Adds a gbrain timeline entry when the document has a date, enabling chronological browsing. - Use Case: After document-interpretation extracts fields from a vendor invoice received via Telegram, this Skill stores it at references/documents/invoices/xyz-inv-2026-001.md with full metadata and a timeline entry. ## Quick Start Store this interpreted invoice with vendor XYZ Sdn Bhd, number INV-2026-001, total RM 12,500, and source telegram://file_id_abc123 to gbrain.

Frequently Asked Questions about document-storage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I store an interpreted document to gbrain?

Pass a JSON object with document_type, fields, summary, and source to the Skill, which calls mcp_gbrain_put_page with the correct path like references/documents/invoices/<vendor>-<number>.md. It also adds a timeline entry when the document has a date.

What document types does gbrain document storage support?

It supports six directory conventions: invoices, quotations, legal, purchase_orders, delivery_orders, and other. Each type has its own filename pattern, such as <vendor_slug>-<invoice_number>.md for invoices.

Does document-storage classify or extract fields from files?

No. Classification and field extraction are handled by document-interpretation, and searching is handled by document-retrieval. This Skill only persists already-structured data to gbrain.

Why is the source field required when storing documents?

The source field (telegram://, gdrive://, or local://) records where the original file came from so the stored record can be traced back. Omitting it breaks traceability to the original document.

What happens if document tags are missing in gbrain?

Missing tags break retrieval. The document tag and the type-specific tag are required in the frontmatter so document-retrieval can find the stored page later.