# Vault Orphans

> Triage orphaned notes (zero in/out wikilinks) in an Obsidian vault. Use when finding orphans, linking them into a MOC, or reconnecting Zettelkasten notes.

- Skill: `laurigates/vault-orphans` (Agent Skill)
- Install (CLI): `npx skillmds@latest add laurigates/vault-orphans`
- Raw SKILL.md: https://api.skillmd.com/api/skills/laurigates/vault-orphans/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: laurigates (https://skillmd.com/u/laurigates)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/laurigates/vault-orphans

---


# Orphan Triage

## When to Use This Skill

| Use this skill when... | Use the alternative instead when... |
|---|---|
| Triaging orphan notes (zero incoming, zero outgoing wikilinks) for archive vs. reconnect | Building a new MOC hub the orphans should link into — use `vault-mocs` |
| Distinguishing expected orphans (inbox, daily notes) from meaningful ones | Discovering the orphan list itself via the running CLI — use `search-discovery` |
| Suggesting archival paths for stale isolated Zettelkasten notes | Classifying or consolidating work-namespace redirect stubs — use `vault-stubs` |

An "orphan" is a note with no incoming wikilinks AND no outgoing wikilinks — disconnected from the knowledge graph. Some orphans are expected; others are the most productive places to add structure.

## Classes of Orphan

| Class | Where | Treat as |
|-------|-------|----------|
| Inbox items | `Inbox/*.md` | Expected; process via `/process-inbox` |
| Daily notes | `Notes/YYYY-MM-DD.md`, `work/notes/…` | Expected; they link out but rarely in |
| Standalone references | `Zettelkasten/*.md` with 0↔0 | **Meaningful** — add linkage |
| Kanban board notes | `Kanban/*.md` | Usually acceptable; boards are self-contained |
| Archive / logs | under `Archive/` subfolders | Expected; stale by design |

The vault-agent graph analyzer classifies each automatically.

## Offline Fallback (App Closed)

The detection methodology above is unchanged — only the **data source** changes when Obsidian (and its `obsidian` CLI / live link index) is closed. The `obsidian` CLI and `vault-agent` analyzers are the **live-index** path; parsing the `.md` corpus directly with Glob/Grep is the **deterministic headless default**, and for batch/scheduled audits it is often the better choice (reproducible, free of app/index state). `vault-frontmatter` already operates this way.

Parse the corpus directly:

- **Frontmatter** — read each note's YAML block between the leading `---` fences; extract `tags`, `aliases`, `context`. See `vault-frontmatter` for YAML-block mechanics.
- **Wikilinks** — match `[[Target]]`, `[[Target|Alias]]`, `[[Target#Heading]]`, `[[folder/Target]]`, and `![[embed]]`. Resolve each target to a note by **basename**, then **relative path**, then **alias** (from frontmatter), all **case-insensitive**. Resolve `![[embed]]` against attachments as well as notes — the attachment folder is per-vault configurable, so read it from `.obsidian/app.json` (`attachmentFolderPath`) and fall back to the vault root / `Files/` only when that key is unset.

Build the link graph from the resolved wikilinks: a note is an orphan when no resolved link targets it (zero incoming) **and** it emits no link that resolves (zero outgoing). The class table above (Inbox / daily / Archive) is path-derived and works identically offline.

## Triage Workflow

For each meaningful orphan:

1. **Read the note** — is the content still relevant, or is this old/dead content?
2. **Identify its primary category** — by tag (e.g., `🛠️/neovim` → Neovim MOC) or by title.
3. **Pick ONE action:**
   - **Link from a MOC** — add `[[Note]]` to the appropriate MOC under the right section
   - **Add an inbound link** from a closely related note
   - **Archive** — move to an `Archive/` subfolder if no longer useful
   - **Delete** — only if empty or entirely superseded

## Never Do

- **Don't add dummy links** like "See also: [[Random]]" just to take the note off the orphan list. That's link pollution.
- **Don't create a new MOC** just to absorb one orphan — see `vault-mocs` for thresholds.
- **Don't assume empty = orphan** — some orphans have substantive content that simply wasn't linked.

## Linking Heuristics

When adding a note to a MOC, match on:

1. **Primary tag category** — a note tagged `🛠️/neovim` belongs in the Neovim MOC.
2. **Content topic** — read the first paragraph; pick the MOC that covers that subject.
3. **Existing cluster** — if notes `A`, `B`, `C` all link to each other but none link from a MOC, add the whole cluster to the MOC under one section heading.

## MOC Section Placement

MOCs typically have sections like `## Core Concepts`, `## Tools`, `## Specific Configurations`. Pick the most specific section that fits; create a new `## Something` section only if 3+ notes fall under the same new heading.

## Batch Pattern

Never modify 100+ MOC links in one commit — that's unreviewable. Use one commit per MOC:

```
feat(mocs): link 12 orphaned CLI tool notes into new CLI Tools MOC
```

## Safety

- If a note is substantive and the user's writing style suggests it was important, lean toward linking rather than deleting.
- If tags are contradictory or missing, link to a broader MOC rather than guessing a specific one.
- Preserve any existing heading structure in the MOC when inserting links.

## Related Skills

- **vault-mocs** — MOC conventions and when to create a new MOC
- **vault-wikilinks** — link syntax and safe-rewrite rules
- **search-discovery** — find-by-tag queries via Obsidian CLI