semantic-query

Translate natural-language game-state questions into RAM predicate evaluations.

Updated Dec 30, 2025
One-click install
npx skills add https://github.com/scawful/afs_scawful --skill semantic-query
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: semantic-query
Source: https://github.com/scawful/afs_scawful/tree/main/skills/semantic-query
Command: npx skills add https://github.com/scawful/afs_scawful --skill semantic-query

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Translates natural-language in-game questions into RAM predicates to quickly determine current conditions and state.

Core Features & Use Cases

  • Context queries like is_overworld, is_dungeon, in_cutscene, and can_control for fast scene assessment.
  • State queries such as is_safe, has <item>, and rupees to verify player status and inventory.
  • Unified interface to map human questions to boolean/value predicates for agent decision making in dynamic game environments.

Quick Start

Use a direct instruction like: Ask semantic-query to evaluate is_overworld and return the result.

Frequently Asked Questions about semantic-query

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I query game state using natural language predicates?

To query game state with natural language predicates, you translate human-readable questions into RAM predicate evaluations on the current memory map. This allows an AI agent to quickly assess conditions like is_overworld or in_cutscene to drive decisions.

How does RAM inspection work for evaluating AI agent game conditions?

RAM inspection for AI agent game conditions works by mapping natural-language questions to boolean or simple value predicates across RAM representations. It evaluates memory addresses to verify player status and inventory, returning simple values to drive agent behavior.

Can I check if a player can_control or has an item by inspecting game memory?

Yes, you can check if a player can_control or has an item by inspecting game memory. The system evaluates state queries such as has <item> and is_safe against the current RAM map to return boolean results for agent decision making.

What is the best way to translate natural-language game questions into boolean values?

The best way to translate natural-language game questions into boolean values is by using a unified interface that maps human queries directly to RAM predicate evaluations. This approach handles scene assessments and inventory checks to return simple values for dynamic environments.

Do I need a memory map to evaluate AI agent predicates in a dynamic game environment?

Yes, you need a current memory map to evaluate AI agent predicates in a dynamic game environment. The system relies on RAM representations to perform predicate evaluations and return boolean results for questions like is_dungeon or is_safe.