pokemon-player

Drive headless Pokemon emulators and read RAM to determine game state.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/HugoAlmeidaMoreira/zeus-agent --skill pokemon-player-hugoalmeidamoreira
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pokemon-player
Source: https://github.com/HugoAlmeidaMoreira/zeus-agent/tree/main/skills/gaming/pokemon-player
Command: npx skills add https://github.com/HugoAlmeidaMoreira/zeus-agent --skill pokemon-player-hugoalmeidamoreira

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Play Pokemon via headless emulation and RAM state reads to enable automated observation and control without a graphical interface.

Core Features & Use Cases

  • Headless gameplay: drive Pokemon games while inspecting RAM-derived state for position, HP, and events.
  • RAM-driven observation: combine with a vision model for real-time dashboards and analysis.
  • Use Case: demonstrate an AI playing Gen 1 games (Red, Blue, Yellow) or FireRed while streaming a dashboard or collecting progress data.

Quick Start

Clone NousResearch/pokemon-agent, create a Python 3.10+ virtual environment, install dependencies with the pyboy extras, place a ROM, and run pokemon-agent serve with the appropriate --rom path and --port.

Frequently Asked Questions about pokemon-player

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

FAQPage Schema
How do I automate Pokemon gameplay using a headless emulator and RAM reads?

Automate Pokemon gameplay by driving a headless emulator and reading RAM to determine game state. This enables AI-assisted training and automated demonstrations by exposing position, HP, and event data through a REST API.

How does RAM-driven observation work for AI agent gameplay?

RAM-driven observation reads emulator memory to extract game state like position and HP without a graphical interface. This state data feeds into a vision analysis workflow to generate real-time dashboards for observable play.

What do I need to set up before running headless Pokemon emulation?

Headless Pokemon emulation requires cloning the pokemon-agent repository, creating a Python 3.10+ virtual environment, installing pyboy dependencies, placing a valid ROM, and running the server with the correct ROM path and port.

Can I use this approach to stream a dashboard for Gen 1 Pokemon games?

Yes, you can stream a live dashboard for Gen 1 games like Red, Blue, and Yellow, as well as FireRed. The setup combines RAM state reads with a vision model to provide real-time analysis and observable automated play.

What REST API endpoints are required for orchestrating AI-driven Pokemon gameplay?

AI-driven Pokemon gameplay requires an active REST API exposing specific endpoints: /health for status, /state for RAM-derived data, /screenshot for visuals, and /action for control inputs to drive the headless emulator.

Are there limitations when using headless emulation without a graphical interface?

Headless emulation without a graphical interface requires a local pokemon-agent setup and valid ROM path. It is limited to supported ROMs like Red, Blue, Yellow, and FireRed, relying entirely on RAM reads and REST API endpoints for control.