Embedding Tools¶
Configuration File: embedding_tools.json
Tool Type: Local
Tools Count: 6
This page contains all tools defined in the embedding_tools.json
configuration file.
Available Tools¶
embedding_database_add (Type: EmbeddingDatabase)¶
Add new documents to an existing embedding database. Generates embeddings for new documents using…
embedding_database_add tool specification
Tool Information:
Name:
embedding_database_add
Type:
EmbeddingDatabase
Description: Add new documents to an existing embedding database. Generates embeddings for new documents using the same model as the original database and appends them to the existing FAISS index.
Parameters:
action
(string) (optional) Action to add documents to existing databasedatabase_name
(string) (required) Name of the existing database to add documents todocuments
(array) (required) List of new document texts to embed and addmetadata
(array) (optional) Optional metadata for each new document (same length as documents)
Example Usage:
query = {
"name": "embedding_database_add",
"arguments": {
"database_name": "example_value",
"documents": ["item1", "item2"]
}
}
result = tu.run(query)
embedding_database_create (Type: EmbeddingDatabase)¶
Create a new embedding database from a collection of documents. Generates embeddings using OpenAI…
embedding_database_create tool specification
Tool Information:
Name:
embedding_database_create
Type:
EmbeddingDatabase
Description: Create a new embedding database from a collection of documents. Generates embeddings using OpenAI or Azure OpenAI models and stores them in a searchable database with FAISS vector index and SQLite metadata storage.
Parameters:
action
(string) (optional) Action to create database from documentsdatabase_name
(string) (required) Name for the new database (must be unique)documents
(array) (required) List of document texts to embed and storemetadata
(array) (optional) Optional metadata for each document (same length as documents)model
(string) (optional) OpenAI/Azure OpenAI embedding model to usedescription
(string) (optional) Optional description for the database
Example Usage:
query = {
"name": "embedding_database_create",
"arguments": {
"database_name": "example_value",
"documents": ["item1", "item2"]
}
}
result = tu.run(query)
embedding_database_load (Type: EmbeddingDatabase)¶
Load an existing embedding database from a local path or external source. Allows importing databa…
embedding_database_load tool specification
Tool Information:
Name:
embedding_database_load
Type:
EmbeddingDatabase
Description: Load an existing embedding database from a local path or external source. Allows importing databases created elsewhere or backed up databases into the current ToolUniverse instance.
Parameters:
action
(string) (optional) Action to load database from external sourcedatabase_path
(string) (required) Path to the existing database directory or filedatabase_name
(string) (required) Local name to assign to the loaded databaseoverwrite
(boolean) (optional) Whether to overwrite existing database with same name
Example Usage:
query = {
"name": "embedding_database_load",
"arguments": {
"database_path": "example_value",
"database_name": "example_value"
}
}
result = tu.run(query)
embedding_database_search (Type: EmbeddingDatabase)¶
Search for semantically similar documents in an embedding database. Uses OpenAI embeddings to con…
embedding_database_search tool specification
Tool Information:
Name:
embedding_database_search
Type:
EmbeddingDatabase
Description: Search for semantically similar documents in an embedding database. Uses OpenAI embeddings to convert query text to vectors and performs similarity search using FAISS with optional metadata filtering.
Parameters:
action
(string) (optional) Action to search the databasedatabase_name
(string) (required) Name of the database to search inquery
(string) (required) Query text to find similar documents fortop_k
(integer) (optional) Number of most similar documents to returnfilters
(object) (optional) Optional metadata filters to apply to search results
Example Usage:
query = {
"name": "embedding_database_search",
"arguments": {
"database_name": "example_value",
"query": "example_value"
}
}
result = tu.run(query)
embedding_sync_download (Type: EmbeddingSync)¶
Download an embedding database from HuggingFace Hub to local storage. Allows accessing databases …
embedding_sync_download tool specification
Tool Information:
Name:
embedding_sync_download
Type:
EmbeddingSync
Description: Download an embedding database from HuggingFace Hub to local storage. Allows accessing databases shared by others or your own backups.
Parameters:
action
(string) (optional) Action to download database from HuggingFacerepository
(string) (required) HuggingFace repository to download from (format: username/repo-name)local_name
(string) (optional) Local name for the downloaded database (optional, defaults to repo name)overwrite
(boolean) (optional) Whether to overwrite existing local database with same name
Example Usage:
query = {
"name": "embedding_sync_download",
"arguments": {
"repository": "example_value"
}
}
result = tu.run(query)
embedding_sync_upload (Type: EmbeddingSync)¶
Upload a local embedding database to HuggingFace Hub for sharing and collaboration. Creates a dat…
embedding_sync_upload tool specification
Tool Information:
Name:
embedding_sync_upload
Type:
EmbeddingSync
Description: Upload a local embedding database to HuggingFace Hub for sharing and collaboration. Creates a dataset repository with the database files and metadata.
Parameters:
action
(string) (optional) Action to upload database to HuggingFacedatabase_name
(string) (required) Name of the local database to uploadrepository
(string) (required) HuggingFace repository name (format: username/repo-name)description
(string) (optional) Description for the HuggingFace datasetprivate
(boolean) (optional) Whether to create a private repositorycommit_message
(string) (optional) Commit message for the upload
Example Usage:
query = {
"name": "embedding_sync_upload",
"arguments": {
"database_name": "example_value",
"repository": "example_value"
}
}
result = tu.run(query)