haiku-duel
HaikuDuel — a 5-round haiku battle between you and R1. Speak your haiku via PTT; R1 generates and judges both haikus using the built-in LLM.
- License 许可
- MIT
- 最近更新
- 2026-09-04
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- haiku-duel
- author
- rabbit-r1-skillz
- version
- 1.1.0
- description
- HaikuDuel — a 5-round haiku battle between you and R1. Speak your haiku via PTT; R1 generates and judges both haikus using the built-in LLM.
02 资源文件 3
- 资产文件card.jpg28.1 KB
- 资产文件icon.png9.0 KB
- 参考文档index.html49.9 KB
03 SKILL.md 正文
HaikuDuel
HaikuDuel is a voice-driven haiku battle on the Rabbit R1. You pick an opponent personality, R1 composes its haiku using the built-in LLM, then you speak yours via PTT. The LLM scores both haikus and announces the round winner. First to 3 round wins (across 5 rounds) takes the match.
Modes
| Mode | Badge | Style |
|---|---|---|
| Warrior | WAR | Intense imagery of steel, wind, and the void |
| Comedian | LOL | Absurdist, snack-themed, existential dread |
| Philosopher | ZEN | Zen paradox, stillness, inspired by Basho/Buson/Issa |
| Chaotic | WTF | Gloriously unhinged word salad that still scans |
| Author Mode | AUT | Channels a specific poet (Basho, Kerouac, Dickinson, Bukowski, Li Bai) |
Controls
| Action | What it does |
|---|---|
| Scroll up / down | Navigate mode select |
| PTT (short press) | Confirm mode / start recording your haiku / re-read scoring |
| PTT hold → release | Stop recording |
| Long press | Open stats / return to main menu |
| Scroll down on end screen | View career stats |
AI Generation
All haiku generation and scoring uses the R1 built-in LLM via PluginMessageHandler.
No external services, no local bridge, no API keys required.
- Haiku generation:
PluginMessageHandler.postMessage({ message, useLLM: true, wantsR1Response: false })— response arrives viawindow.onPluginMessagewithdata.messagecontaining the three lines. - Line-by-line speech: each haiku line is spoken using
PluginMessageHandlerwithuseLLM: false, wantsR1Response: true. - Scoring: the same LLM call returns a JSON object with
scoreA,scoreB,winnerLabel, andcommentary.
Storage Keys
| Key | Contents |
|---|---|
haikuduel:stats |
Career stats object (wins, losses, draws, rounds, haikusSpoken, winsByMode, lossesByMode) |
haikuduel:best-haiku |
All-time best haiku { text, score, mode } |
haikuduel:match-history |
Last 20 match records |
Setup
No setup required. HaikuDuel runs entirely on-device using the R1's built-in LLM.
04 同仓库更多 Skills
whisper-quiz
WhisperQuiz — voice-answered trivia game for Rabbit R1. Generates questions via the R1's built-in LLM, captures spoken answers via STT, and evaluates correctness with flexible LLM-based matching.
walk-cast
Walking podcast production skill. Receives raw audio metadata and transcript from the R1, removes long pauses, applies noise reduction and normalization, generates AI chapter titles and episode metadata via LLM, and uploads the final episode to your configured podcast RSS host (Spotify for Podcasters / Anchor, Podbean, Buzzsprout, RSS.com, or any FTP-accessible feed).
voice-novel
Local machine skill for VoiceNovel — handles story export to disk.
voice-guide
Environment description and navigation assistant for blind and low-vision users. Analyzes camera frames via a vision LLM, reads visible text, identifies scene categories (people/vehicles/food/hazards/exits), performs 360° scans, retrieves GPS position, and sends emergency SMS alerts — all via audio feedback with zero screen dependency.
voice-draft
Voice-driven document drafting for the Rabbit R1. Dictate stream-of-consciousness speech; the R1 formats it into Email, Blog Post, Report, Proposal, Meeting Notes, Creative Writing, Code Comments, Cover Letter, or Tweet Thread — then saves the result to a configured folder on your local machine via OpenClaw.
voice-commit
Hands-free git and CI workflow voice controller for the Rabbit R1. Speak git/CI commands and hear results — commit, push, status, PR management, test runs, and deploy, all voice-driven via OpenClaw.