Cakir Commander

If you have any questions, please feel free to reach out to me at the email address below.

Mail: [email protected]

Prompt-to-Action AI Framework: Dynamic Behavior Synthesis

Warning: This framework currently supports Windows (x64) environments. Requires manual setup of external inference binaries as per the provided documentation.

Overview

This asset is a high-level Prompt-to-Action Framework designed to bridge the gap between natural language processing and deterministic game logic. It provides a robust pipeline to transform unstructured text prompts into executable Dynamic Behavior Trees at runtime, allowing for unprecedented player agency and NPC flexibility.

Unlike static AI systems, this framework treats game methods as modular nodes that can be synthesized into complex logic chains on the fly.

Technical Core Features

Dynamic Behavior Synthesis: Automatically constructs and executes behavior trees at runtime based on parsed intent.

Method Registration System: A streamlined API to register existing C# methods, making them "visible" and "actionable" to the logic parser.

DSL (Domain Specific Language) Integration: Translates natural language into a structured logic format (Cakir DSL), ensuring game-safe execution without the risks of direct code injection.

Runtime Logic Mapping: Real-time mapping of user intent to specific game functions like navigation, combat, or environmental interaction.

Workflow Integration

Register: Map your existing game functions (e.g., MoveTo, PerformAction, CheckCondition) to the framework's registry.

Parse: Feed natural language input into the system (e.g., "Patrol the area, but if you see an intruder, sound the alarm and retreat").

Execute: The framework generates a temporary, optimized Behavior Tree and executes the logic within the Unity lifecycle.

Developer Control & Customization

This framework is built for developers who want to implement "Instruction-Based" gameplay. It provides the architectural "plumbing" needed to handle complex command strings, allowing you to focus on game design rather than the intricacies of intent parsing and logic sequencing.