Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Cross-Linker (Obsidian Auto Cross-Referencing)

Scans an Obsidian vault, scores missing [[wikilinks]] by confidence, and actually writes them into your pages.

UtilitiesIntermediate★ 3,520⑂ 350AI score 8/10Last updated: Oct 3, 2026

What it does

  • Builds a page registry by grepping frontmatter only (title, aliases, tags, category) across every .md file in the vault.
  • Detects unlinked mentions of page names, aliases and concepts, plus semantic signals like shared tags, same-project proximity and peripheral→hub reach (with case- and diacritic-insensitive matching).
  • Scores each candidate and labels it EXTRACTED / INFERRED / AMBIGUOUS, acting only on the first two.
  • Inserts inline links at the first natural mention, or falls back to a ## Related section when no natural anchor exists.
  • Infers a typed relationship (uses, extends, contradicts, derived_from, related_to, …) and records it in the page's relationships: frontmatter.
  • Computes project affinity for misc/ pages and flags promotion candidates.
  • Takes a Git snapshot commit before writing (only if the vault is its own repo), then emits a table report and syncs the wiki log; optionally refreshes a QMD search index.

Who it's for

  • Obsidian power users whose vault keeps growing but whose graph stays sparse and full of orphans.
  • Anyone already running the llm-wiki framework and obsidian-wiki CLI from this repo.
  • People who ingest large batches of notes and want new pages woven into the existing knowledge graph automatically.

Example uses

  1. "Link my pages" → full scan, 23 links added across 12 pages, with a confidence-labelled report.
  2. Run right after a bulk ingest → freshly imported notes get connected to existing concept and entity pages.
  3. "My wiki feels disconnected" → orphan pages are identified, plausible targets proposed and linked, remaining orphans listed.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/Ar9av/obsidian-wiki/HEAD/.skills/cross-linker/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the .skills/cross-linker folder from the GitHub repo Ar9av/obsidian-wiki into my ~/.claude/skills/cross-linker/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/Ar9av/obsidian-wiki.git && mkdir -p ~/.claude/skills && cp -r obsidian-wiki/.skills/* ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repo: git clone https://github.com/Ar9av/obsidian-wiki.git
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the skills in: cp -r obsidian-wiki/.skills/* ~/.claude/skills/
    • Copy all of them — this skill depends on the companion llm-wiki skill.
  5. Configure your vault: set OBSIDIAN_VAULT_PATH=/path/to/vault in a .env file or the global config, and optionally OBSIDIAN_LINK_FORMAT=wikilink (or markdown).
  6. (Recommended) Run git init and make one commit inside the vault so the skill can take a rollback snapshot before writing.
  7. Restart Claude Code and ask: "find missing links and cross-reference my wiki".
  8. Because this skill writes to your notes, try it on a test vault or a backup copy first and review the report before trusting INFERRED links.
View source on GitHub ↗License: MIT