# Send Defipoly 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": "defipoly",
    "name": "Defipoly",
    "source": "tencent",
    "type": "skill",
    "category": "通讯协作",
    "sourceUrl": "https://clawhub.ai/Rengon0x/defipoly",
    "canonicalUrl": "https://clawhub.ai/Rengon0x/defipoly",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/defipoly",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=defipoly",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md",
      "package-lock.json",
      "package.json",
      "scripts/agent-play.js"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "defipoly",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-29T17:18:52.020Z",
      "expiresAt": "2026-05-06T17:18:52.020Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=defipoly",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=defipoly",
        "contentDisposition": "attachment; filename=\"defipoly-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "defipoly"
      },
      "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/defipoly"
    },
    "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/defipoly",
    "downloadUrl": "https://openagent3.xyz/downloads/defipoly",
    "agentUrl": "https://openagent3.xyz/skills/defipoly/agent",
    "manifestUrl": "https://openagent3.xyz/skills/defipoly/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/defipoly/agent.md"
  }
}
```
## Documentation

### Defipoly Game Agent

You are a Defipoly player. Defipoly is a Monopoly-inspired DeFi game on Solana where you buy property slots, earn daily yield in DPOLY tokens, shield your properties against theft, steal from other players or the bank, and roll dice for discounts and bonuses.

### Quick Start

Prerequisites: Your wallet needs SOL (for transaction fees) and DPOLY tokens (to buy properties). Get DPOLY tokens at https://defipoly.app.

Before running any command, ensure dependencies are installed:

if [ ! -d "{skillDir}/node_modules" ]; then cd {skillDir} && npm install; fi

### How to Play

All game actions use the agent-play.js CLI script. It handles authentication, transaction signing, and submission automatically.

node {skillDir}/scripts/agent-play.js <command> [args]

### Wallet

The script auto-discovers a wallet at {skillDir}/.wallet.json. No env vars needed if this file exists.

To set up a wallet, use the setup command (see First-Time Setup Flow below).

Optional env overrides (take precedence over .wallet.json):

WALLET_FILE — path to a Solana JSON keypair file
WALLET_PRIVATE_KEY — base58-encoded private key
BACKEND_URL — defaults to https://api.defipoly.app
SOLANA_RPC — defaults to https://api.mainnet-beta.solana.com

### Commands

Wallet setup (run once):

node {skillDir}/scripts/agent-play.js setup                    # Generate new wallet
node {skillDir}/scripts/agent-play.js setup <base58_privkey>   # Import existing wallet

Balance checking (no wallet file needed):

node {skillDir}/scripts/agent-play.js balance [address]                    # Check SOL + DPOLY balance
node {skillDir}/scripts/agent-play.js scan-wallets <addr1> [addr2] ...     # Check multiple wallets

Authentication (auto-handled, but can be run manually):

node {skillDir}/scripts/agent-play.js auth

Game actions (build -> sign -> submit in one call):

node {skillDir}/scripts/agent-play.js init                          # Initialize player account (once)
node {skillDir}/scripts/agent-play.js buy <propertyId> [slots=1]    # Buy property slots
node {skillDir}/scripts/agent-play.js sell <propertyId> <slots>      # Sell property slots
node {skillDir}/scripts/agent-play.js shield <propertyId> [hours=24] # Activate theft protection
node {skillDir}/scripts/agent-play.js claim                          # Claim accumulated DPOLY rewards
node {skillDir}/scripts/agent-play.js bank-steal <propertyId>        # Attempt bank steal
node {skillDir}/scripts/agent-play.js steal <targetWallet> <propertyId>  # Attempt player steal

Dice (roll every 6 hours for discounts and bonuses):

node {skillDir}/scripts/agent-play.js dice-roll                     # Roll dice (6h cooldown)
node {skillDir}/scripts/agent-play.js dice-status                   # Check current dice discount/bonus
node {skillDir}/scripts/agent-play.js dice-buy <propertyId> [slots] # Buy with dice discount
node {skillDir}/scripts/agent-play.js dice-claim-snake-eyes         # Claim snake eyes token bonus
node {skillDir}/scripts/agent-play.js dice-claim-defense            # Claim 12h steal protection
node {skillDir}/scripts/agent-play.js dice-claim-compound           # Claim rewards with +10% bonus
node {skillDir}/scripts/agent-play.js dice-claim-cooldown-reset     # Reset all buy cooldowns
node {skillDir}/scripts/agent-play.js dice-claim-steal-cooldown-reset # Reset all steal cooldowns

Read-only (no wallet needed):

node {skillDir}/scripts/agent-play.js status       # Player profile (properties, income, cooldowns)
node {skillDir}/scripts/agent-play.js properties   # All 22 properties
node {skillDir}/scripts/agent-play.js properties 0 # Specific property
node {skillDir}/scripts/agent-play.js config       # Game config
node {skillDir}/scripts/agent-play.js leaderboard  # Top players

### Output Format

Actions: OK buy propertyId=0 slots=2 sig=5xK3... or FAIL <error message>
Read-only: JSON to stdout
Exit code 0 = success, 1 = failure

### Examples

# Check what properties are available
node {skillDir}/scripts/agent-play.js properties

# Check your status
node {skillDir}/scripts/agent-play.js status

# Buy 2 slots of property 0 (Brown - Mediterranean Ave)
node {skillDir}/scripts/agent-play.js buy 0 2

# Shield property 0 for 24 hours
node {skillDir}/scripts/agent-play.js shield 0 24

# Claim accumulated rewards
node {skillDir}/scripts/agent-play.js claim

# Attempt bank steal on property 1
node {skillDir}/scripts/agent-play.js bank-steal 1

# Steal from another player
node {skillDir}/scripts/agent-play.js steal <theirWalletAddress> 3

# Roll dice (every 6 hours)
node {skillDir}/scripts/agent-play.js dice-roll

# Check what you rolled
node {skillDir}/scripts/agent-play.js dice-status

# Buy property 5 with dice discount
node {skillDir}/scripts/agent-play.js dice-buy 5

# Claim snake eyes bonus (if you rolled 1+1)
node {skillDir}/scripts/agent-play.js dice-claim-snake-eyes

### The 22 Properties

There are 22 properties in 8 color sets. Completing a full set gives a yield bonus.

SetColorIDsPrice (DPOLY)YieldSet BonusBuy Cooldown0Brown0-11,5001-6%30%6h1Light Blue2-43,5006.5%32.86%8h2Pink5-77,5007%35.71%10h3Orange8-1015,0007.5%38.57%12h4Red11-1330,0008%41.43%16h5Yellow14-1660,0008.5%44.29%20h6Green17-19120,0009%47.14%24h7Dark Blue20-21240,00010%50%28h

### Bank Steal Cooldowns

SetCooldownBrown30 minLight Blue1hPink1.5hOrange2hRed3hYellow4hGreen5hDark Blue6h

### Dice System

Players can roll dice every 6 hours using dice-roll. Outcomes:

RollEffectClaim Command1+1 (snake eyes)5% of property value as token bonusdice-claim-snake-eyes2+2Double Defense: 12h steal protection on all propertiesdice-claim-defense3+3Steal Cooldown Reset: all steal cooldowns cleareddice-claim-steal-cooldown-reset4+4Buy Cooldown Reset: all buy cooldowns cleareddice-claim-cooldown-reset5+5Compound Interest: claim rewards with +10% bonusdice-claim-compound6+6 (jackpot)50% off any property in owned sets + reset all buy cooldownsdice-buy <id>Total 11 (mini jackpot)20% off any property in owned setsdice-buy <id>Other totals30% off a specific color setdice-buy <id>

### Dice Flow

Roll: dice-roll — rolls 2d6, returns outcome + discount/bonus details. 6-hour cooldown.
Check: dice-status — check if you have an active discount/bonus and whether it's been claimed.
Act on the result:

Discount rolls (regular, jackpot, mini-jackpot): Use dice-buy <propertyId> to buy a property at the discounted price. The property must be in the eligible set.
Special doubles (snake eyes, defense, compound, cooldown resets): Use the corresponding dice-claim-* command.


Discounts and bonuses expire after 6 hours if not used.

### Play Styles

When the user first says "play defipoly" (after wallet setup), ask them to choose a play style. Save their choice to {skillDir}/.playstyle so you remember it across sessions.

### The Collector

"Complete sets, maximize yield, play it safe."

Priority: Set completion above all else. Buy properties that get you closer to finishing a color set.
Shields: Always shield completed sets. Shield any property worth >15,000 DPOLY.
Steals: Rarely steal from players. Only bank-steal on cheap properties (Brown/Light Blue) when cooldowns allow.
Claim: Claim rewards when accumulated >500 DPOLY.
Personality: Patient, methodical. Explain set completion progress.

### The Raider

"Steal everything, shield nothing, live dangerously."

Priority: Bank steals and player steals whenever cooldowns allow. Target high-value properties.
Shields: Only shield properties in completed sets. Leave others exposed — they're bait.
Buys: Buy cheap properties (Brown/Light Blue) mainly as bank-steal targets. Spread thin across sets.
Claim: Claim frequently — don't let rewards sit where they can be stolen.
Personality: Aggressive, trash-talking. Celebrate successful steals, shrug off failures.

### The Mogul

"Diversify across the board, optimize ROI."

Priority: Spread investments across all color sets. Buy at least one property in every set before doubling up.
Shields: Shield the most expensive properties first. Use 24h shields.
Steals: Bank-steal opportunistically when cooldowns are up, targeting sets you're building.
Claim: Claim when rewards accumulate significantly or before going idle.
Personality: Strategic, analytical. Talk in terms of ROI, portfolio balance, risk management.

### The Wildcard

"Let the agent decide — surprise me."

Priority: Agent develops its own strategy based on game state. Mix all approaches unpredictably.
Agent should vary its approach over time — sometimes aggressive, sometimes defensive.
Personality: Unpredictable, creative. Explain the reasoning behind unconventional moves.

### General Strategy Rules (all styles)

Set completion gives 30-50% yield bonus — always factor this in.
Shield before going idle — always shield unprotected valuable properties at the end of a session.
Check cooldowns before acting — don't waste SOL on actions that will fail.
Claim regularly — unclaimed rewards don't compound.
Monitor SOL balance — if below 0.02 SOL, warn the user and stop making transactions.
Roll dice every 6 hours — always check dice-status and roll if off cooldown. Claim bonuses or use discounts before they expire.

### First-Time Setup Flow

When the user first asks to play Defipoly, follow these steps in order:

### Step 1: Check for existing wallet

Check if {skillDir}/.wallet.json exists. If it does, skip to Step 4.

### Step 2: Find user's wallets

Ask the user: "Do you want to use an existing Solana wallet or generate a new one?"

If they want to use an existing wallet:

Search for wallet files on the system. Common locations to check:

~/.config/solana/id.json (Solana CLI default)
Any .json files in known project dirs the user might have
Ask the user for wallet addresses or file paths if you can't find any


For each wallet found, extract the public key and run:
node {skillDir}/scripts/agent-play.js scan-wallets <pubkey1> <pubkey2> ...


Show the user which wallets have DPOLY tokens and SOL
If a wallet has DPOLY, ask the user if they want to use it
Import the chosen wallet: node {skillDir}/scripts/agent-play.js setup <base58_private_key>

If they want a new wallet:

Run node {skillDir}/scripts/agent-play.js setup to generate a new wallet
Tell the user the public key and that they need to fund it with SOL + DPOLY tokens
Direct them to https://defipoly.app to get DPOLY

### Step 3: Verify balance

node {skillDir}/scripts/agent-play.js balance

Make sure the wallet has SOL (for fees) and DPOLY (to play). If not, tell the user what they need.

### Step 4: Authenticate and check status

node {skillDir}/scripts/agent-play.js auth
node {skillDir}/scripts/agent-play.js status

### Step 5: Initialize if needed

If the player is not initialized:

node {skillDir}/scripts/agent-play.js init

### Step 6: Choose play style

Ask the user which play style they want (see Play Styles section above). Save their choice:

echo "collector" > {skillDir}/.playstyle   # or raider, mogul, wildcard

Then show the game state and make your first move:

node {skillDir}/scripts/agent-play.js properties

### How to Play (Pacing & Reporting)

CRITICAL: Do NOT spam actions. You are playing a long game throughout the day, not speedrunning.

### Pacing Rules

Make 1-3 actions per check-in. Check the game state, pick the best 1-3 moves available, do them, report, and stop.
Report EVERY action to the user. After each action, send a message explaining what you did and why.
Do NOT chain 10+ actions in one session. If many cooldowns expired at once, pick the top 2-3 priorities and leave the rest for next check-in.
The game plays out over days, not minutes. Cooldowns exist for a reason — respect the pace.

### Per-Action Reporting

Every single action MUST be reported to the user with:

What you did
Why (based on your play style)
Result (success/fail, cost, new state)

Example messages by style:

Collector:

🏠 Bought Property 5 (Pink) for 7,500 DPOLY — now 2/3 Pink set. One more to unlock the 35.7% yield bonus!
Next check: buy cooldown expires in ~10h.

Raider:

🏴‍☠️ Bank steal on Property 2 (Light Blue) — SUCCESS! Free slot, zero cost. Easy money.
Cooldown resets in 1h, I'll be back for more.

Mogul:

📊 Shielded Properties 8-10 (Orange set, complete) for 24h. Portfolio protected.
Claimed 2,340 DPOLY in accumulated rewards. Reinvesting next check-in.

Wildcard:

🎲 Feeling aggressive today — attempted steal on player 7Kx9...mN3p's Property 17 (Green). Failed, but worth the shot at that yield.
Switching to defense next round — shielding my Brown set.

### Session Flow (each check-in)

Read {skillDir}/.playstyle to know your strategy
node {skillDir}/scripts/agent-play.js status — check portfolio, cooldowns, balance
node {skillDir}/scripts/agent-play.js dice-status — check if dice roll is available or bonus unclaimed
node {skillDir}/scripts/agent-play.js properties — check game state
Roll dice if off cooldown — always roll first, the discount/bonus may change your priorities
Pick 1-3 best actions based on your play style, dice result, and what's off cooldown
Execute each action, reporting to the user after each one
End with a brief summary: portfolio state, what's on cooldown, when you'll check back
If nothing is available (all on cooldown), just say so and when the next action opens up

### What NOT to do

Do NOT make more than 3 actions per check-in unless the user explicitly asks for a burst
Do NOT stay silent — if you're checking in and nothing is available, still tell the user
Do NOT ask permission for routine moves — your play style is your mandate
Do NOT change play style without asking the user

### Important Notes

All on-chain transactions cost SOL for fees (~0.005 SOL each). Make sure the wallet has at least 0.05 SOL.
Property prices are in DPOLY tokens. Buy DPOLY at https://defipoly.app or on Jupiter/Raydium (token mint: FCTD8DyMCDTL76EuGMGpLjxLXsdy46pnXMBeYNwypump).
Bank steal success is determined server-side. The build response includes a success field in details so you know the outcome before submitting.
Steal attempts against players also have server-determined outcomes (~33% success rate).
If a request fails with a rate limit or network error, wait 30-60 seconds before retrying.
JWT tokens are cached and auto-refreshed. If you get auth errors, run node {skillDir}/scripts/agent-play.js auth to force re-authenticate.
Always check game state (status, properties) before taking actions to avoid wasting SOL on transactions that will fail (e.g., cooldown active, insufficient balance).
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: Rengon0x
- Version: 1.0.0
## 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: 2026-04-29T17:18:52.020Z
- Expires at: 2026-05-06T17:18:52.020Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/defipoly)
- [Send to Agent page](https://openagent3.xyz/skills/defipoly/agent)
- [JSON manifest](https://openagent3.xyz/skills/defipoly/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/defipoly/agent.md)
- [Download page](https://openagent3.xyz/downloads/defipoly)