What does this skill do, and when should you use it?
This skill is specifically for organizing and maintaining Obsidian vaults, and is only activated when the user explicitly mentions Obsidian. It can create Maps of Content (MOCs), add or fix wikilinks, find orphan notes, normalize YAML frontmatter, generate Dataview/Bases dashboards, and organize folders and deduplicate notes. All operations respect Obsidian's core conventions.
Reads note content, searches for [[ ]] links and orphan notes, uses write_file to create MOCs and dashboards, and edit_block to edit note content (insert wikilinks, normalize frontmatter). It does not move or rename files, instead instructing the user to do so inside Obsidian.
- An Obsidian user wants to build a navigation structure for a topic and needs MOCs created with linked notes.
- A user notices notes are poorly linked and wants to identify and fix orphan notes automatically.
- A user wants consistent frontmatter across notes to filter with Dataview or Bases.
- A user wants a dashboard listing recently updated notes.
- A user wants to merge duplicate notes and fix all inbound links to the discarded note.
How do you install this skill?
- The skill relies on Desktop Commander MCP, which has broad permissions like terminal execution; the skill itself does not restrict its scope, so users should be aware of security boundaries and use it only in trusted environments.
- Although the skill recommends performing moves/renames inside Obsidian, actual execution might still use MCP file operations; users should understand the risks.
- The skill file is in English, which may impose translation or comprehension costs for Chinese users.
- Static review cannot verify reliability; it is recommended to test on a small vault before actual use.
- Local filesystem
Place the folder containing plugins/claude/skills/obsidian-vault/SKILL.md into your Agent Skills directory (e.g., ~/.claude/skills/ for Claude Code).
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/wonderwhy-er/DesktopCommanderMCP.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/claude/skills/obsidian-vault" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Help me organize my Obsidian vault and find all orphan notes.
Mention Obsidian directly in your request, for example: 'Help me organize my Obsidian vault and find all orphan notes.' The skill will first confirm the vault root and Dataview plugin availability, then proceed.
What are this skill's strengths and limitations?
- Follows Obsidian best practices (wikilinks, frontmatter, MOCs)
- Provides concrete Dataview queries and MOC templates
- Explicitly guides to avoid link breakage
- Works with any Agent Skills-compatible client, no extra dependencies
- Relies on user cooperation for moves/renames inside Obsidian
- Requires Dataview plugin for its full features
- No test suite or verified examples provided
- Search may be slow on very large vaults
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Obsidian Vault Organizer this page | 49 · Use with care | ★ 10k | 4d ago | MIT |
| Obsidian Wiki Health Check | 48 · Use with care | ★ 15k | 1mo ago | MIT |
| Obsidian Markdown Syntax Assistant | 45 · Not recommended | ★ 15k | 1mo ago | MIT |
| Obsidian Markdown Editor | 57 · Use with care | ★ 49k | 26d ago | MIT |
| Compound Vault Methodology Modes | 56 · Use with care | ★ 15k | 1mo ago | MIT |
How did FollowSkills review this skill?
The skill file explicitly recommends conservative operations: only edit content via edit_block/write_file, never use move_file for moves/renames (which would break wikilinks), and perform moves/renames inside Obsidian. This reflects least privilege and avoidance of destructive operations. However, there is no explicit user confirmation mechanism (the skill itself doesn't provide a confirmation step), and the skill relies on Desktop Commander MCP which has broad permissions like terminal access, but the skill file does not restrict its scope. Also, no rollback or backup strategy is provided. Therefore, deduct points and give 16.
The skill instructions are detailed, but static review cannot execute and verify the reliability of key workflows (e.g., start_search, edit_block). The skill file provides multiple methods (e.g., Dataview queries), but no test evidence or clear error handling is provided. The behavior of the underlying MCP tools is unverified in this review. Due to static calibration limits, reliability maximum is 10, but lack of test coverage and failure feedback leads to a score of 5.
Trigger conditions are very clear: only use when user explicitly mentions Obsidian, with concrete examples and a clear non-fit scope. The audience is AI agents. In terms of environment fit, Obsidian features do not rely on overseas services, but the skill file is in English and does not provide Chinese support, potentially affecting usability for Chinese users, so deduct points. Give 10.
The skill has clear name, version (0.1.0), and audience; good information architecture; clear sections; provides templates and examples; and has explicit known limitations (e.g., renames must go through Obsidian). However, installation/dependency notes are missing (as a skill running in an MCP environment, dependencies are not documented), and there is no clear maintenance responsibility or update path. License is explicit (MIT), but the skill file itself does not include license information. Therefore, give 9.
The skill's goals are clear, steps are detailed, and expected outputs are explicit (e.g., MOC template, frontmatter example). But static review cannot verify actual effectiveness, and no representative outputs are provided to prove direct usability. According to static calibration, effectiveness maximum is 7, but lack of empirical evidence gives a score of 5.
The repository contains test files (e.g., test/ab-test.test.js, test/integration/edit-block-performance.js) and a CI workflow (codespell.yml), but these tests target the MCP itself, not the key paths of this skill. The skill file itself provides no reproducible verification examples or cross-source evidence. Therefore, give 4.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →