# Send Snake Rodeo to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "snake-rodeo",
    "name": "Snake Rodeo",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/okwme/snake-rodeo",
    "canonicalUrl": "https://clawhub.ai/okwme/snake-rodeo",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/snake-rodeo",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=snake-rodeo",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md",
      "clawdhub.json",
      "daemon/autoplay.mjs",
      "lib/api.mjs",
      "lib/config.mjs",
      "lib/process.mjs"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "snake-rodeo",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-03T19:10:35.803Z",
      "expiresAt": "2026-05-10T19:10:35.803Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=snake-rodeo",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=snake-rodeo",
        "contentDisposition": "attachment; filename=\"snake-rodeo-3.2.1.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "snake-rodeo"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/snake-rodeo"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/snake-rodeo",
    "downloadUrl": "https://openagent3.xyz/downloads/snake-rodeo",
    "agentUrl": "https://openagent3.xyz/skills/snake-rodeo/agent",
    "manifestUrl": "https://openagent3.xyz/skills/snake-rodeo/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/snake-rodeo/agent.md"
  }
}
```
## Documentation

### Snake Rodeo Skill

Play the Trifle Snake Rodeo automatically with a persistent daemon and modular strategy system. Built on the snake-rodeo-agents library.

### How It Works

The game is a multiplayer snake on a grid (hex or cartesian). Teams bid on directions each round — the highest bidder's direction wins. All bids go into a prize pool that the winning team splits. The daemon watches the game via SSE, picks optimal directions using a strategy, and submits votes automatically.

### Prerequisites

Authenticated via trifle-auth skill
Node.js 18+
Ball balance (earned from games, auth bonuses, etc.)

### Daemon Commands

# Start/stop
node snake.mjs start [--detach] [--strategy NAME]
node snake.mjs stop
node snake.mjs status
node snake.mjs attach [-f]

# Pause/resume voting (daemon keeps running)
node snake.mjs pause
node snake.mjs resume

# Configuration
node snake.mjs config [key] [value]
node snake.mjs strategies
node snake.mjs server [live|staging]
node snake.mjs telegram [chat_id|off]

# Manual play
node snake.mjs state
node snake.mjs vote <direction> <team> [amount]
node snake.mjs strategy    # Analyze current game
node snake.mjs balance

### Strategies

Five built-in strategies are available. Each extends BaseStrategy from snake-rodeo-agents.

StrategyAliasDescriptionexpected-valueev, defaultBFS pathfinding, dead-end avoidance, game-theoretic team selection, probabilistic defection in multi-agent scenarios. Balanced.aggressiveaggBacks leading teams, counter-bids aggressively.underdogundBacks small pools for bigger payouts.conservativeconMinimum bids, prioritizes safety.randomrandRandom valid moves.

Switch strategy:

node snake.mjs config strategy aggressive
# or
node snake.mjs start --strategy aggressive

### Creating Custom Strategies

Extend BaseStrategy from snake-rodeo-agents:

import { BaseStrategy } from 'snake-rodeo-agents';

export class MyStrategy extends BaseStrategy {
  constructor(options = {}) {
    super('my-strategy', 'My custom strategy', options);
  }

  computeVote(parsed, balance, state) {
    // parsed: ParsedGameState — hex grid, teams, scores, valid directions
    // balance: number — current ball balance
    // state: AgentState — round tracking, team assignment, vote history

    // Return a vote:
    return { direction: 'ne', team: 'A', amount: 1, reason: 'chasing fruit' };

    // Or skip:
    return { skip: true, reason: 'too risky' };
  }

  // Optional: counter-bid when outbid
  shouldCounterBid(parsed, balance, state, ourVote) {
    return null; // or return a new VoteAction
  }
}

Key types for strategy development:

ParsedGameState — Parsed game with head, teams[], validDirections[], gridRadius, prizePool, minBid, fruitsToWin
AgentState — { currentTeam, roundSpend, roundVoteCount, lastRound, gamesPlayed, votesPlaced, wins }
VoteAction — { direction, team, amount, reason }

### snake-rodeo-agents Library

The core logic lives in snake-rodeo-agents, a standalone TypeScript library. This skill wraps it with daemon management, config persistence, and OpenClaw integration.

### API Client

import { SnakeClient, createAndAuthenticate, parseGameState, getStrategy } from 'snake-rodeo-agents';

// Auth (creates a throwaway wallet, no real ETH needed)
const { token, privateKey, address } = await createAndAuthenticate('https://bot.trifle.life');

// Create client
const client = new SnakeClient('https://bot.trifle.life', token);

// Play
const rawState = await client.getGameState();
const parsed = parseGameState(rawState);
const strategy = getStrategy('expected-value');
const vote = strategy.computeVote(parsed, balance, agentState);

if (vote && !vote.skip) {
  await client.submitVote(vote.direction, vote.team, vote.amount);
}

SnakeClient methods:

MethodDescriptiongetGameState()Current game state (snake, fruits, scores, votes)getBalance()Current ball balancesubmitVote(dir, team, amount)Submit a direction votegetRodeos()List active rodeo gamesgetUserStatus()User profile and stats

### Wallet Authentication

SIWE (Sign In With Ethereum) auth with throwaway wallets:

import { createAndAuthenticate, reauthenticate, checkToken } from 'snake-rodeo-agents';

// New wallet
const { token, privateKey, address } = await createAndAuthenticate(serverUrl);

// Reuse saved wallet
const { token } = await reauthenticate(serverUrl, savedPrivateKey);

// Check token validity
const user = await checkToken(serverUrl, token);

### Game State Utilities

The library provides hex grid utilities for strategy development:

import { parseGameState, hexDistance, bfsDistance, floodFillSize, getValidDirections } from 'snake-rodeo-agents';

const parsed = parseGameState(rawState);

// BFS shortest path to a target (respects snake body, grid bounds)
const { distance, firstDir } = bfsDistance(head, target, rawState);

// Flood-fill reachable area (dead-end detection)
const reachable = floodFillSize(head, rawState);

// Hex distance between two positions
const dist = hexDistance(posA, posB);

### Tournament Simulator

Run offline tournaments to compare strategies at high speed:

# CLI
npm run simulate -- ev,aggressive --games 100 --seed 42
npm run simulate -- ev,aggressive,conservative --config small --verbose
npm run simulate -- ev,aggressive --json   # machine-readable output

// Library
import { SimAgent, runTournament, RODEO_CYCLES, getStrategy } from 'snake-rodeo-agents';

const agents = [
  new SimAgent('a', 'ev-agent', getStrategy('ev')),
  new SimAgent('b', 'agg-agent', getStrategy('aggressive')),
];

const results = runTournament(agents, RODEO_CYCLES, 100, { seed: 42 });
console.log(results.agentStats);
// Same seed = identical results for reproducibility

Simulator options:

FlagDescription-g, --games NGames per config (default: 100)-c, --config NAMEsmall|medium|large|all (default: all)-s, --seed NRNG seed for reproducibility-v, --verbosePrint per-round details--jsonMachine-readable JSON output

### Telegram Logging

Send game events to a Telegram group:

import { TelegramLogger } from 'snake-rodeo-agents';

const tg = new TelegramLogger({
  botToken: process.env.TELEGRAM_BOT_TOKEN,
  chatId: process.env.TELEGRAM_CHAT_ID,
});

await tg.send('<b>Hello</b> from the snake agent!');

Configure in the daemon:

node snake.mjs telegram <chat_id>   # enable
node snake.mjs telegram off          # disable

### Configuration

Settings are stored in ~/.config/snake-rodeo/settings.json (XDG-compliant, isolated from any host agent).

KeyDefaultDescriptionstrategyexpected-valueActive strategy nameserverlivelive or stagingminBalance5Minimum balance to place votestelegramChatIdnullTelegram chat ID for loggingtelegramBotTokennullTelegram bot token (or set TELEGRAM_BOT_TOKEN env var)

### File Locations

PurposePathSettings~/.config/snake-rodeo/settings.jsonAuth token~/.config/snake-rodeo/auth.json or TRIFLE_AUTH_TOKEN env varDaemon state~/.local/state/snake-rodeo/daemon.stateDaemon PID~/.local/state/snake-rodeo/daemon.pidDaemon log~/.local/share/snake-rodeo/daemon.log

### Authentication

The skill resolves your Trifle auth token in this order:

TRIFLE_AUTH_TOKEN environment variable (recommended for automation)
~/.config/snake-rodeo/auth.json — { "token": "your-jwt-here" }

To set up auth, run snake auth login (uses trifle-auth skill) or set the env var directly.

### Architecture

snake-game/                             # OpenClaw skill wrapper
├── SKILL.md                            # This file
├── snake.mjs                           # CLI entry point
├── clawdhub.json                       # ClawHub registry metadata
├── package.json                        # Dependencies (snake-rodeo-agents)
├── lib/
│   ├── config.mjs                      # Settings/paths
│   ├── api.mjs                         # Token-based API (uses OpenClaw auth)
│   ├── process.mjs                     # Daemon PID management
│   └── telegram.mjs                    # Telegram bridge
├── daemon/
│   └── autoplay.mjs                    # Game loop: SSE → strategy → vote
└── node_modules/
    └── snake-rodeo-agents/             # Core library (TypeScript)
        └── dist/
            ├── lib/game-state.js       # Hex grid, BFS, flood-fill
            ├── lib/strategies/         # Strategy implementations
            ├── lib/client.js           # Standalone API client
            ├── lib/auth.js             # Wallet SIWE auth
            ├── lib/simulator.js        # Offline game simulator
            ├── lib/telegram.js         # Telegram logger
            └── bin/play.js             # Standalone CLI runner

### Upgrading

node snake.mjs stop
cd ~/.openclaw/workspace/skills/snake-rodeo
npm install github:trifle-labs/snake-rodeo-agents
node snake.mjs start --detach
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: okwme
- Version: 3.2.1
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-03T19:10:35.803Z
- Expires at: 2026-05-10T19:10:35.803Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/snake-rodeo)
- [Send to Agent page](https://openagent3.xyz/skills/snake-rodeo/agent)
- [JSON manifest](https://openagent3.xyz/skills/snake-rodeo/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/snake-rodeo/agent.md)
- [Download page](https://openagent3.xyz/downloads/snake-rodeo)