LLM Workflow Generation From Natural Language for Security Playbooks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Automated workflows in computing environments, particularly in cybersecurity settings, require programming skills for definition and can be complex and resource-intensive, leading to inefficiencies and errors.

Innovation Solution

Utilizing large language models (LLMs) to generate workflow definitions from natural language descriptions, enabling users without programming skills to create accurate and efficient workflows through structured plan generation and pseudocode representation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If workflow definitions are written in programming language, then precision and control are improved, but ease of operation deteriorates

Engineering Contradiction:
Improveworkflow definition precisionVSAvoidease of workflow creation
Core Design Contradiction:
Manufacturing precisionVSEase of operation

Solution Approach 1:

The patent introduces an intermediary system that translates natural language descriptions into workflow definition language. This intermediary acts as a bridge between users who lack programming skills and the precise workflow definition language, automatically converting casual language into structured workflow definitions without requiring users to learn programming.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical system of manual programming with an automated translation system. Instead of users manually writing workflow definition language, the system automatically generates precise workflow definitions from natural language inputs, substituting human cognitive effort with automated processing.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Ease of operation

If GUI-based tools are used for constructing workflows, then ease of operation is improved, but productivity deteriorates

Engineering Contradiction:
Improveease of workflow creationVSAvoidworkflow creation speed
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent replaces the manual drag-and-drop GUI interaction with automated text processing. The system processes natural language descriptions and automatically generates complete workflow definitions, eliminating the time-consuming manual assembly process while maintaining ease of use through natural language input.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables users to create workflows through simple text descriptions without requiring interaction with complex GUI elements. The automated translation system handles the entire workflow creation process, allowing users to service themselves by simply describing what they need rather than learning and using complex tools.

Inventive Principle:
Principle #25Self-service

3Ease of operation

If manual workflow construction is performed, then ease of operation is improved, but loss of time increases

Engineering Contradiction:
Improveworkflow creation accessibilityVSAvoidworkflow definition time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The patent substitutes manual workflow construction with automated text-to-workflow translation. The system processes natural language descriptions and instantly generates workflow definitions, reducing the time required from minutes or hours of manual assembly to seconds of automated processing while maintaining accessibility for non-programmers.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250348360A1Automated workflow creation
Publication Date: 2025.11.13 MICROSOFT TECHNOLOGY LICENSING LLC
  • US20250348360A1 patent drawing
  • US20250348360A1 patent drawing
  • US20250348360A1 patent drawing

AI summary

A computer-implemented method generates workflow definitions in workflow definition language using one or more large language models, LLMs. The method includes receiving a natural language description of an automated workflow and generating a plan generation prompt including the natural language description and plan generation instructions. The plan generation prompt is input to one of the LLMs and in response a structured plan comprising a plurality of actions are received. For each action, a corresponding segment of workflow definition language is generated to provide a plurality of segments of workflow definition language. The segments are combined to form a workflow definition corresponding to the natural language description.