Schema Infer

Schema Inference

Auto-detect metadata conventions from your existing notes. Zero configuration - the system learns from what you've already written.

When to Use

Invoke when you want to:

• Understand what frontmatter fields are standard in a folder
• Discover implicit schemas from existing notes
• Find required vs optional fields
• See naming patterns for files

Arguments

Process

1. Parse User Input

Recognize schema inference requests:

• "what conventions does my meetings folder use?"
• "infer schema for projects/"
• "what fields should be in daily notes?"

2. Call MCP Tool

mcp__flywheel__infer_folder_conventions({
  folder: "meetings/",
  min_confidence: 0.5
})

3. Format Results

## Inferred Schema: meetings/

Based on 47 notes in this folder:

### Required Fields (>90% presence)
| Field | Type | Common Values |
|-------|------|---------------|
| type | string | "meeting" (100%) |
| date | date | - |
| attendees | array[wikilink] | - |

### Optional Fields (50-90% presence)
| Field | Type | Common Values |
|-------|------|---------------|
| status | string | scheduled (40%), completed (50%), cancelled (10%) |
| tags | array | meeting, standup, review |

### Suggested Computed Fields
- word_count (not currently tracked)
- link_count (not currently tracked)

### Naming Convention
Files follow pattern: `YYYY-MM-DD *.md`

### Coverage
- 94% of notes have frontmatter
- 3 notes may be incomplete (see /schema-gaps)

Use Cases

• New folder setup: "What schema should my new project folder follow?"
• Documentation: "Document the metadata conventions for this vault"
• Consistency check: "Are my meeting notes following a pattern?"
• Onboarding: "What fields does this team use?"

Integration

Works well with other skills:

• schema-gaps: Find notes missing expected fields
• schema-apply: Apply inferred conventions to incomplete notes
• vault-schema: See vault-wide field overview

Version: 1.0.0