vault-ingest

SkillDocs & knowledge

Lets your agent sort raw content into the right knowledge base locations and link related entries.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the vault-ingest skill

About this capability

Classify raw content and route it to the appropriate knowledge base location

What this skill tells your AI

The instructions your AI receives, as published by pass-agent/loomkin in .agents/skills/vault-ingest/SKILL.md and read by ahel’s review.

Analyze raw input and route to the right location in the knowledge base.

Classify Content

Determine what this is:

  • Strategy, process, or how-to: note (link to parent topic)
  • Hub or organizing content: topic
  • Product or brand information: project
  • Person information: person
  • A decision or commitment: decision (also log to decision graph)
  • External reference: source
  • A feature or product idea: idea

If the input contains multiple distinct concepts, extract each as a separate atomic entry.

Check for Duplicates

vault_search(query: "{key phrases from input}") — if a similar entry exists, suggest updating it instead of creating a new one. Use ask_user to confirm.

Create Entries

vault_create_entry(entry_type: "{classified_type}", ...) for each extracted concept.

For decisions, also create a decision graph node: decision_log(node_type: "decision", title: "...", confidence: ...)

Link

  • Link new notes to parent topics via vault_link(link_type: "parent")
  • Link related entries via vault_link(link_type: "related")
  • Extract action items to vault_kanban(action: "add", ...)

Report: files created with paths, topics linked, action items added.

Signals

GitHub stars
180
Forks
28
Last commit
May 2026
Advanced
Catalog kind
skill
Gateway key
vault-ingest
Source
github.com/pass-agent/loomkin