/catalog

Context Catalog

File a document into the context vault with proper frontmatter and wikilinks to relevant project overviews.

Workflow

1. Read the source document

If the argument is a URL (starts with http):

• Use the obsidian:defuddle skill to extract clean markdown if available
• Otherwise use WebFetch to read the page content
• Proceed with the extracted content as the document to catalog

If the argument is a file path, read the file directly.

Determine its nature:

• Client document (sent by client, external)
• Deliverable (produced by us for the client)
• Research (technical research, reference material)

2. Determine filing location

Category	Location
Client docs	context-vault/docs/client-docs/{slug}.md
Deliverables	context-vault/docs/deliverables/{slug}.md
Research	context-vault/docs/research/{slug}.md

3. Propose filing

Present: category, destination path, frontmatter, which project overview(s) to link from. Wait for approval.

4. File the document (on approval)

Text files (.md, .txt): Write to destination with frontmatter (title, source, cataloged date, category, tags) followed by content.

Binary files (.pdf, .docx, images): Copy original to destination. Write a markdown companion with frontmatter + text summary.

5. Link from project overviews

Add the document as a wikilink in the Project Documents section of relevant overview(s).

6. Report

Confirm what was filed and where it was linked.

Hard Rules

1. Do NOT extract entities — that's /ctx:ingest's job
2. Always ask for approval before writing
3. Always link from at least one project overview