Fix Broken Vault Links

VerifiedSafe

Scans an Obsidian vault for broken wikilinks, categorizes them by type (missing notes, typos, case mismatches, moved notes), and offers automatic or manual fixes. Helps maintain link integrity and avoid dead references.

Sby Skills Guide Bot
ContentIntermediate
6/2/2026
Claude Code
#vault#wikilinks#broken-links#link-repair#autonomous-fixing

Recommended for

Automated summary

Scans and automatically repairs broken wikilinks in an Obsidian-style vault, categorizing issues and using confidence scores for autonomous fixes.

Strengths

  • Accurate classification of broken link types (missing notes, typos, case mismatches, moved notes)
  • Confidence scoring to prioritize auto-fixes vs. user review
  • Partial autonomy with user validation before applying changes

Limitations

  • Requires a specific MCP tool (flywheel) not available in all environments
  • Only handles wikilinks, not external links or text references
  • Moved note detection reliability depends on fuzzy title matching
When to use it

When your vault has many broken links after renaming or deleting notes and you want a guided, semi-automated repair process.

When not to use it

If you prefer manual review of each broken link or if your setup lacks the required MCP tool.

Security analysis

Safe
Quality score90/100

The skill only invokes a read-only MCP tool to find broken wikilinks and presents a preview. It requires explicit user consent to spawn an autonomous repair agent, and does not itself execute any destructive or network actions.

No concerns found

How this analysis is made

Examples

Fix all broken wikilinks
Fix the broken links in my vault. Show me a preview of all broken wikilinks and then repair them autonomously.
Check link health
Check for broken wikilinks in my vault and give me a breakdown by type (missing notes, typos, case mismatches, moved/renamed).
Repair specific folder
Find and repair broken links only inside the 'Projects' folder, showing me a preview first.

name: vault-fix-links description: Find and repair broken wikilinks in vault. Triggers when user mentions "fix links", "broken links", "repair vault", "fix broken links". auto_trigger: true trigger_keywords:

  • "fix links"
  • "broken links"
  • "repair vault"
  • "fix broken links"
  • "repair links"
  • "find broken links"
  • "broken wikilinks"
  • "fix wikilinks"
  • "invalid links"
  • "dead links"
  • "link errors"
  • "repair connections"
  • "link problems"
  • "fix references"
  • "bad links" allowed-tools: mcp__flywheel__find_broken_links, Task, TodoWrite

Fix Broken Links

Preview broken wikilinks and spawn autonomous repair agent.

When to Use

Invoke when you want to:

  • Find all broken wikilinks in vault
  • Preview what's broken and why
  • Repair broken links automatically
  • Improve vault link health

Process

1. Find Broken Links

Call mcp__flywheel__find_broken_links to retrieve all broken wikilinks.

2. Analyze Break Patterns

Categorize broken links by type:

Missing Notes (create candidates):

  • [[New Topic]] → No file exists
  • High confidence if referenced multiple times
  • Suggest creating note

Typos (fuzzy match):

  • [[Databrics]] → Did you mean [[Databricks]]?
  • Use Levenshtein distance
  • Auto-fix if confidence >90%

Case Mismatches (exact match different case):

  • [[databricks]] → Exists as [[Databricks]]
  • Auto-fix (safe)

Moved Notes (search by title):

  • [[Old Path/Note]] → Now at [[New Path/Note]]
  • Search vault for matching title
  • Suggest update

3. Show Preview

Display first 20 broken links with categorization:

Broken Links Analysis
═══════════════════════════════════════════════
Found 200 broken links across 50 notes

📊 Breakdown by Type:
  • Missing Notes: 80 (40%)
  • Typos: 40 (20%)
  • Case Mismatches: 30 (15%)
  • Moved/Renamed: 50 (25%)

🔍 Sample (showing 20 of 200):

Missing Notes:
1. [[API Guide]] (15 references)
   → Mentioned in: Projects, Tech Docs
   → Suggestion: Create tech/guides/API.md

Typos:
2. [[Databrics]] → [[Databricks]] (confidence: 95%)
   → Auto-fixable

Case Mismatches:
3. [[azure]] → [[Azure]] (exact match)
   → Auto-fixable

═══════════════════════════════════════════════

💡 Recommendations:
  • Auto-fix: 50 links (high confidence >90%)
  • User review: 60 links (medium confidence 50-90%)
  • Manual fix: 90 links (low confidence <50%)

🤖 Spawn autonomous repair agent?
  • Will process all 200 broken links
  • Auto-fix high confidence (>90%)
  • Present choices for medium confidence (50-90%)
  • Skip low confidence (<50%)

Type 'yes' to spawn link-repair agent

4. User Decision

Ask user if they want to:

  • Option A: Spawn agent for autonomous fixing
  • Option B: See full list (export to note)
  • Option C: Fix specific folder only
  • Option D: Cancel (just wanted to see the damage)

Confidence Scoring

Typo Detection (Levenshtein Distance):
- Distance 1: 98% confidence
- Distance 2: 85% confidence
- Distance 3+: <70% confidence

Case Mismatch:
- Exact title match: 100% confidence
- Always safe to auto-fix

Missing Note Analysis:
- Referenced 20+ times: High priority (suggest create)
- Referenced 5-19 times: Medium priority
- Referenced 1-4 times: Low priority (might be scratch)

Moved Note Detection:
- Title exact match: 95% confidence
- Title fuzzy match: 60-85% confidence
- Search by content similarity: <60% confidence

Output Format

Always use the branded format:

Broken Links Analysis
═══════════════════════════════════════════════

[Analysis content]

═══════════════════════════════════════════════

Six Gates Compliance

| Gate | Implementation | |------|----------------| | 1. Read Before Write | Finds all broken links via MCP before any fixes | | 2. File Exists | Validates target notes exist for link suggestions | | 3. Chain Validation | Agent verifies each fix before proceeding | | 4. Mutation Confirmation | Shows preview, requires explicit "yes" to proceed | | 5. Health Check | Uses MCP find_broken_links for vault access | | 6. Post Validation | Agent reports what was fixed after completion |

Safety

  • Non-destructive: Only suggests, doesn't auto-apply
  • User approval: Requires explicit "yes" to spawn agent
  • Preview first: Always show what's broken before fixing
  • Backup: Agent will backup before batch edits

Performance

  • Discovery: 5-10 seconds (find_broken_links MCP call)
  • Analysis: 10-20 seconds (categorize links)
  • Preview: Instant (show first 20)
  • Total: ~30 seconds before agent spawn decision

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/fix-broken-vault-links-mm5t8b85/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/fix-broken-vault-links-mm5t8b85' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/fix-broken-vault-links-mm5t8b85/download' --output '.skill-downloads/fix-broken-vault-links-mm5t8b85/SKILL.md.part' && mv '.skill-downloads/fix-broken-vault-links-mm5t8b85/SKILL.md.part' '.skill-downloads/fix-broken-vault-links-mm5t8b85/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

Tools declared by the author (check availability in your agent): mcp__flywheel__find_broken_links, Task, TodoWrite

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/fix-broken-vault-links-mm5t8b85/SKILL.md' && test -s '.skill-downloads/fix-broken-vault-links-mm5t8b85/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/vault-fix-links' && cp -R '.skill-downloads/fix-broken-vault-links-mm5t8b85' '.claude/skills/vault-fix-links'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/fix-broken-vault-links-mm5t8b85'

openskills read 'fix-broken-vault-links-mm5t8b85'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/fix-broken-vault-links-mm5t8b85/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills