AnyLlmSearchDocument
Full text search of a LLM document for a text pattern (regex) for more results as a companion to similarity search that returns a few top results. Currently supports only war room file entries, search results, and text that has been preprocessed in XSOAR prior to uploading to the LLM. (See AnyLlmUploadText, AnyLlmUploadFileEntry and AnyLlmUploadDocument). Results are placed in the search results buffer for where they can be added to the LLM's conversation context.
python · Anything LLM
Details
| ID | AnyLlmSearchDocument |
|---|---|
| Language | python |
| From Version | 6.10.0 |
| Docker Image | demisto/python3:3.12.8.3296088 |
README
Full text search of a LLM document for a text pattern (regex) for more results as a companion to similarity search that returns a few top results. Currently supports only war room file entries, search results, and text that has been preprocessed in XSOAR prior to uploading to the LLM. (See AnyLlmUploadText, AnyLlmUploadFileEntry and AnyLlmUploadDocument). Results are placed in the search results buffer for where they can be added to the LLM’s conversation context
Script Data
| Name | Description |
|---|---|
| Script Type | python3 |
Inputs
| Argument Name | Description |
|---|---|
| title | title of the LLM document to search with the XSOAR file entry ID. Example: “37@25496_anythingllm.txt” |
| pattern | regex to search text for [ see re.findall() ] |
Outputs
There are no outputs for this script.
args: - description: 'title of the LLM document to search with the XSOAR file entry ID. Example: "37@25496_anythingllm.txt".' name: title required: true - description: regex to search text for [ see re.findall() ]. name: pattern required: true comment: Full text search of a LLM document for a text pattern (regex) for more results as a companion to similarity search that returns a few top results. Currently supports only war room file entries, search results, and text that has been preprocessed in XSOAR prior to uploading to the LLM. (See AnyLlmUploadText, AnyLlmUploadFileEntry and AnyLlmUploadDocument). Results are placed in the search results buffer for where they can be added to the LLM's conversation context. commonfields: id: AnyLlmSearchDocument version: -1 contentitemexportablefields: contentitemfields: fromServerVersion: '' dockerimage: demisto/python3:3.12.8.3296088 enabled: true engineinfo: {} name: AnyLlmSearchDocument runas: DBotWeakRole runonce: false script: '' scripttarget: 0 subtype: python3 tags: [] type: python fromversion: 6.10.0 tests: - No tests (auto formatted)