"Guides MaaNTE audio, rhythm, MIDI, Tetris, BagelSpam, and
Scanned 9/8/2026
Install to Claude Code
npx -y skills add VectorSpaceLab/AREX-Skill --skill media-minigames --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Media Minigames?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/vectorspacelab-media-minigames)More formats (shields.io, HTML) on the badges page.
---
name: media-minigames
description: "Guides MaaNTE audio, rhythm, MIDI, Tetris, BagelSpam, and
AutoFScroll automation workflows."
disable-model-invocation: true
metadata:
disco-role: operating
license: AGPL 3.0
---
# Media and Minigames
## Use This When
Use this sub-skill for MaaNTE features where audio, music, rhythm, or lightweight minigame automation is the core task:
- `SoundDodge` audio dodge/counter.
- `Rhythm` / auto rhythm play and select-song logic.
- `AutoPiano` MIDI playback.
- `Tetris` AI and vitality handling.
- `BagelSpam` photo/text/LLM post generation.
- `AutoFScroll` quick pickup.
For general gameplay menus or task options, return to [../gameplay-tasks/SKILL.md](../gameplay-tasks/SKILL.md). For underlying Python action mechanics, return to [../custom-actions/SKILL.md](../custom-actions/SKILL.md).
## Read First
1. [references/media-workflows.md](references/media-workflows.md) for task behavior, options, and key implementation facts.
2. [references/troubleshooting.md](references/troubleshooting.md) for audio, MIDI, Windows-key, and minigame-specific failures.
3. [../../references/task-catalog.md](../../references/task-catalog.md) for the task table and option names.
4. Use the bundled helper scripts when you need safe file/input checks:
- `scripts/check_midi_file.py`
- `../../scripts/check_maante_environment.py --summary`
## General Notes
- Audio workflows are sensitive to host services and devices. A Linux import check does not prove PulseAudio/PipeWire capture or Windows loopback behavior.
- Rhythm and Tetris live loops should respect `context.tasker.stopping` and avoid unbounded sleep loops without stop checks.
- AutoPiano uses Windows keyboard message injection in the bridge layer; it is not a pure cross-platform library workflow.
- BagelSpam may run in preset mode or LLM mode. LLM mode depends on an OpenAI-compatible HTTP endpoint and API key; keep that dependency explicit.
- AutoFScroll is a simple holding/scrolling helper, but its usefulness depends on the user already holding F to activate the action in-game.
## Validation
Safe checks:
```bash
python scripts/check_midi_file.py path/to/file.mid
python ../../scripts/check_maante_environment.py --summary
```
Live verification requires the appropriate game mode/window and should be short, bounded, and user-approved. Use a tiny song, one rhythm round, one Tetris round, one BagelSpam post generation, or one audio threshold check rather than a long farm loop.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!