Lola Semantic StoreInstall

Overview

The Lola Semantic Store node allows you to interact with a semantic text storage system within n8n. It supports two main operations:

• Save: Store a text (optionally split into chunks) in a named collection for later semantic search.
• Query: Search the semantic store for texts similar to a given query, retrieving the closest matches.

This node is useful in scenarios where you need to:

• Build a knowledge base or FAQ system that can be semantically searched.
• Store and retrieve documents based on meaning rather than exact keywords.
• Implement AI-powered search features in your workflows.

Example use cases:

• Saving customer support tickets and later querying for similar issues.
• Storing product descriptions and enabling semantic product search.
• Creating a searchable archive of research papers or articles.

---

Properties

Name	Meaning
Operation	Selects the action: • Save – Store a text in the semantic store • Query – Query the semantic store
Text	The text to be stored in the semantic store (used when Operation is "Save")
Collection Name	The name of the collection to use for storing or querying data
Text to Query	The text to search for similar entries in the semantic store (used when Operation is "Query")
K	Number of top results to return from the query (used when Operation is "Query")
Treshold	Maximum distance to be considered a match (used when Operation is "Query")
Document ID	Identifier for the document being saved (used when Operation is "Save")
Chunk Size	Size in tokens for splitting the text into chunks (used when Operation is "Save")
Overlap	Overlap in tokens between chunks when splitting the text (used when Operation is "Save")

---

Output

• The node adds or updates a field called lola_smenatic_store in the output item's json object.
• For Save operation:
  • lola_smenatic_store.collectionName: The name of the collection where the text was saved.
• For Query operation:
  • lola_smenatic_store[Collection Name]: The result(s) of the semantic query, structure depends on the underlying implementation but typically includes matched documents and their similarity scores.

Example output (for Save):

{
  "lola_smenatic_store": {
    "collectionName": "my_collection"
  }
}

Example output (for Query):

{
  "lola_smenatic_store": {
    "my_collection": [
      {
        "documentId": "doc1",
        "score": 0.12,
        "text": "..."
      },
      ...
    ]
  }
}

Note: Actual fields inside the query result may vary depending on the backend.

---

Dependencies

• May require credentials for lolaKeyApi if configured.
• Relies on external functions (saveSemanticStore, querySemanticStore) which may depend on an external semantic store service.
• No mandatory API key unless specified in your n8n instance.

---

Troubleshooting

• Missing Required Fields:
  If required properties (e.g., Collection Name, Document ID, etc.) are not provided, the node will throw an error indicating missing parameters.
• Connection/Service Errors:
  Errors thrown by the underlying semantic store service (e.g., network issues, authentication failures) will be surfaced as node errors.
• Invalid Query Parameters:
  Setting inappropriate values for K or Treshold may result in empty or unexpected results.

Common error messages:

• "Missing parameter: Collection Name" – Ensure all required fields are filled.
• "NodeOperationError: ..." – Indicates an issue with the semantic store service or invalid input; check the error details for more information.

---

Links and References

• n8n Documentation
• Semantic Search (Wikipedia)
• For further integration details, consult your Lola Semantic Store provider's documentation.