Vector Notes Indexer

Description

Create a simple local embedding index for notes using vector embeddings. Enables semantic search and similarity matching for note collections.

Inputs

{
  "notes": ["Note texts to index"],
  "model": "embedding model name",
  "index_path": "Path to save index"
}

Outputs

{
  "ok": true,
  "data": {
    "indexed": 10,
    "index_size": "5.2MB"
  }
}

Usage

Demo Mode

python scripts/main.py --demo

Create Index

echo '{"notes":["Note 1","Note 2"],"model":"sentence-transformers"}' | python3 scripts/main.py

Examples

Example 1: Index Notes

$ echo '{"notes":["Meeting notes about Q1 planning","Project status update"],"model":"sentence-transformers","index_path":"./notes_index"}' | python3 scripts/main.py
{
  "ok": true,
  "data": {
    "indexed": 2,
    "index_size": "1.2MB"
  }
}

Error Handling

When an error occurs, the skill returns:

{
  "ok": false,
  "error": "Error description",
  "details": {
    "notes": "Notes list is required"
  }
}