Connection Strength Calculator

Calculate the overall relationship quality between two notes using multiple factors.

When to Use

Invoke when you want to:

• Quantify how related two notes are
• Understand relationship factors (not just links)
• Compare relationship strengths between different pairs
• Get a holistic view beyond simple link counting

Arguments

Process

1. Parse User Input

Identify the two notes:

• "connection strength between [[Project A]] and [[Technology X]]"
• "how related are [[React]] and [[Vue]]"
• "relationship score for daily note and weekly note"

2. Call MCP Tool

mcp__flywheel__get_connection_strength(
  note_a: "path/to/first/note.md",
  note_b: "path/to/second/note.md"
)

3. Format Results

Connection Strength
=================================================

[[Project A]] <-> [[Technology X]]

Overall Score: 78/100 (Strong Connection)

-------------------------------------------------

Factor Breakdown:

  Direct Links:        25/30  (+25)
    - A links to X: Yes (10 pts)
    - X links to A: Yes (10 pts)
    - Link frequency: 3 mentions (5 pts)

  Shared References:   18/25  (+18)
    - Common neighbors: 6 notes
    - Overlap ratio: 72%

  Structural:          20/25  (+20)
    - Same folder: Yes (10 pts)
    - Path distance: 1 hop (10 pts)

  Metadata:            15/20  (+15)
    - Shared tags: 3 (#project, #tech, #active)
    - Same type: Yes (project notes)

-------------------------------------------------

Interpretation:
  - STRONG: These notes are highly related
  - Direct bidirectional links exist
  - High shared context (6 common references)
  - Same folder reinforces relationship
  - Shared tags confirm topical similarity

=================================================

Score Interpretation

Scoring Factors

The strength score considers:

1. Direct Links (up to 30 points)

   • Does A link to B?
   • Does B link to A?
   • How many times are they mentioned?

2. Shared References (up to 25 points)

   • Common neighbors count
   • Overlap ratio with total links

3. Structural (up to 25 points)

   • Same folder = higher connection
   • Shorter path = higher connection

4. Metadata (up to 20 points)

   • Shared tags
   • Same note type (if using frontmatter)

Use Cases

• Relationship validation: "Is this connection as strong as I thought?"
• Comparison: "Which project is more related to this technology?"
• Graph optimization: "Should I add more links between these?"
• Knowledge mapping: "What's the strongest connection in this cluster?"

Integration

Works well with other skills:

• path: Trace the actual link path between notes
• common: See the specific shared references
• bidirectional: Check if direct two-way link exists
• clusters: See which cluster each note belongs to

Version: 1.0.0