Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Polymarket Oracle

Multi-strategy arbitrage and trading bot for Polymarket prediction markets. Scans ALL markets (crypto, politics, sports, economics, entertainment) for parity...

MIT-0 · Free to use, modify, and redistribute. No attribution required.
0 · 269 · 0 current installs · 0 all-time installs
byWesley Armando@georges91560
MIT-0
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
high confidence
Purpose & Capability
The name/description (Polymarket arbitrage bot) match the included code and runtime behavior (scanning Polymarket APIs and placing orders). Requiring POLYMARKET_API_KEY/SECRET/PASSPHRASE is appropriate for trading. However the registry metadata claimed no required env/configs while SKILL.md and code require API credentials — a mismatch in metadata vs implementation.
!
Instruction Scope
Runtime instructions and docs explicitly instruct reading/writing credentials files (e.g., /etc/polymarket-oracle/credentials.env), creating API keys using a wallet private key, and show a systemd unit that embeds secrets. The docs both warn against storing the WALLET_PRIVATE_KEY on the server and simultaneously provide a systemd example that places WALLET_PRIVATE_KEY in Environment lines (contradiction). The instructions therefore allow (and even encourage, via the service example) storing highly sensitive secrets on the running host — scope creep beyond a scanner-only role.
Install Mechanism
There is no automated install spec (instruction-only install), and the code uses only the Python standard library at runtime. The README/config suggest optionally installing py-clob-client locally to create API keys; this is reasonable. No arbitrary network-installs or downloads from suspicious hosts are present in the manifest. Risk is primarily operational (how user deploys), not from a packaged installer.
!
Credentials
Requesting Polymarket API_KEY/SECRET/PASSPHRASE is proportionate to trading. However the presence of WALLET_PRIVATE_KEY in multiple places (env docs, systemd service example, credentials file examples) is problematic: the project alternately says 'only use private key locally once' and then shows ways to put the private key on the server/service. Registry metadata omitted required env vars entirely, increasing confusion. The skill thus mixes appropriate credential requests with instructions that could expose full-wallet private keys — disproportionate if the intent is only to run trades via API keys.
Persistence & Privilege
The skill does not request always:true and does not modify other skills. The documentation strongly encourages running the bot as a systemd service (enable on boot, auto-restart) which increases persistence on a host — normal for a trading bot. Combined with the unsafe secret-handling examples, persistent deployment increases the blast radius if secrets are stored insecurely.
What to consider before installing
This package implements a real-money trading bot and needs your Polymarket API key/secret/passphrase — that part is expected. Key concerns: (1) The manifest metadata is inconsistent with the SKILL.md/code (metadata claims no env vars but the code requires credentials). (2) The docs both tell you NOT to store your WALLET_PRIVATE_KEY on the server and then provide a systemd example that embeds it into the service file/credentials file — do not follow that example. Best practices before installing: generate Polymarket API keys locally using your wallet private key, then provide only the API_KEY/SECRET/PASSPHRASE to the running bot; never put your wallet private key on the server or in systemd Environment lines. Run the bot in simulation mode first; review the full code yourself (or have a trusted reviewer) to confirm it only uses the API keys for trading and does not exfiltrate credentials. If you proceed, store credentials in a tightly permissioned EnvironmentFile (chmod 600 root:root), avoid embedding secrets in unit files, and consider running within an isolated VM/container with restricted network access and monitoring. If you want greater assurance, ask the author to fix the manifest (declare required env vars and remove contradictory examples) and to remove any examples that recommend embedding private keys in service units.

Like a lobster shell, security has layers — review code before you run it.

Current versionv1.0.1
Download zip
arbitragevk97abzybykhrtybky3m7t7ppgs81zs9yautomationvk97abzybykhrtybky3m7t7ppgs81zs9ylatestvk97eagavs8dc8vghhx01cy1dxx81yy1tmulti-strategyvk97abzybykhrtybky3m7t7ppgs81zs9yopportunitiesvk97abzybykhrtybky3m7t7ppgs81zs9ypolymarketvk97abzybykhrtybky3m7t7ppgs81zs9yprediction-marketsvk97abzybykhrtybky3m7t7ppgs81zs9yscannervk97abzybykhrtybky3m7t7ppgs81zs9ytradingvk97abzybykhrtybky3m7t7ppgs81zs9y

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Runtime requirements

🎯 Clawdis
Binspython3

SKILL.md

Polymarket Oracle — Multi-Strategy Arbitrage Bot

🎯 WHAT IT DOES

Scans ALL Polymarket markets for profitable opportunities.

Markets covered:

  • Crypto (BTC, ETH, altcoins - 5min, 15min, daily markets)
  • Politics (Elections, policy, polls, debates)
  • Sports (NBA, NFL, MLB, NHL, Soccer, UFC, Tennis, Golf)
  • Economics (Fed rates, CPI, jobs, GDP, stock markets)
  • Technology (Apple, Tesla, Google, IPOs, launches)
  • Entertainment (Oscars, Emmys, box office, celebrities)
  • Weather (Hurricanes, temperatures, climate events)
  • Miscellaneous (Viral events, trends, binary outcomes)

Total: 1000-5000+ markets scanned continuously


🔥 STRATEGIES DETECTED

1. Parity Arbitrage

What: YES + NO prices don't sum to $1.00

YES price: $0.45
NO price: $0.52
Total: $0.97

Buy both → Guaranteed payout: $1.00
Profit: $0.03 (3.1%)

Risk: ZERO (guaranteed profit)

Frequency: 5-20 per day
Duration: 2-15 seconds
Win rate: 100% (if executed)


2. Logical Arbitrage

What: Impossible price combinations between related markets

Market A: "Chiefs win Super Bowl" = 28%
Market B: "AFC team wins" = 24%

Impossible! Chiefs ARE an AFC team!

Buy "AFC wins" at 24%
If Chiefs win (28%) → AFC wins → Profit
If other AFC wins → Profit
Cannot lose unless NFC wins

Frequency: 2-10 per day
Duration: Minutes to hours
Win rate: 90-95%


3. Tail-End Trading

What: Buy highly certain outcomes (>95%) just before resolution

Event: Fed keeps rates (99% certain)
Market price: $0.97

Buy at $0.97
Resolution: $1.00
Profit: $0.03 (3.1%)

Risk: Very low (event almost certain)

Frequency: 10-30 per day
Duration: Hours to days
Win rate: 95-98%


4. Market Making

What: Provide liquidity, earn spread + maker rebates

Token midpoint: $0.50

Place orders:
BID: $0.475 (maker)
ASK: $0.525 (maker)

Spread capture: $0.05 (10%)
+ Polymarket rebates
+ Liquidity rewards (3x multiplier)

Monthly: 10-25% annualized

Frequency: Continuous
Duration: Indefinite
Win rate: 80-90% (inventory risk)


5. Latency Arbitrage

What: Trade on information before market adjusts

Bitcoin 5-min market:
- Monitor Binance/Chainlink feed
- When price crosses threshold
- Bot knows resolution 2-15s before UI

Trade before market reacts

Frequency: 50-200 per day (crypto markets)
Duration: Seconds
Win rate: 70-85%


6. Combinatorial Arbitrage (AI-powered)

What: Find mispriced relationships between multiple markets

Using sentence embeddings (e5-large-v2)
Find related markets:
- "Trump wins election"
- "Trump wins popular vote"
- "Trump wins by 5%+"

If Trump wins by 5%+ → wins popular vote
But prices show 30% vs 48% = Mispricing

Frequency: 3-15 per day
Duration: Hours to days
Win rate: 75-90%


📊 MARKET CATEGORIZATION

Automatic categorization by keywords:

Categories = {
    'crypto': Bitcoin, Ethereum, DeFi, altcoins
    'politics': Elections, presidents, votes, polls
    'sports': NBA, NFL, Soccer, UFC, Tennis
    'economics': Fed, CPI, jobs, GDP, stocks
    'technology': Apple, Tesla, Google, IPOs
    'entertainment': Oscars, movies, celebrities
    'weather': Hurricanes, temperatures, climate
    'other': Everything else
}

Strategy application:

  • Crypto → Latency + Market Making
  • Politics → Logical + Tail-End
  • Sports → Parity + News Arbitrage
  • Economics → Tail-End + Market Making
  • All → Parity + Combinatorial

⚡ EXECUTION SPEED

Parallel Scanning:

Workers: 50 parallel threads
Markets scanned: 1000-2000 per minute
Opportunities detected: Real-time
Execution: <1 second from detection

Scan Cycle:

00s - Fetch all active markets (Gamma API)
05s - Categorize markets automatically
10s - Parallel scan with 50 workers
30s - Detect opportunities across 5 strategies
35s - Alert via Telegram
40s - Wait 20 seconds
60s - REPEAT

💰 CAPITAL ALLOCATION

Default configuration:

Total capital: $10,000 (configurable)
Max per market: $1,000 (1-10% of capital)

Strategy allocation:
- Parity arb: 20% ($2,000)
- Tail-end: 40% ($4,000)
- Market making: 30% ($3,000)
- Latency: 10% ($1,000)

Position sizing:

  • High certainty (>95%): $500-$5,000
  • Medium certainty (90-95%): $200-$2,000
  • Speculative (<90%): $100-$1,000

📱 TELEGRAM ALERTS

Real-time notifications:

🎯 POLYMARKET OPPORTUNITIES

Found 27 opportunities:

PARITY_ARBITRAGE: 8
TAIL_END: 12
MARKET_MAKING: 5
LATENCY: 2

Top 5 by profit:

• parity_arbitrage: 4.23%
  NBA Lakers vs Warriors - Game Winner...

• tail_end: 3.87%
  Fed Rate Decision - Keeps Rates...

• parity_arbitrage: 3.54%
  Bitcoin Closes Above $45K Today...

• tail_end: 3.12%
  Super Bowl Winner - Chiefs...

• market_making: 2.89%
  Ethereum Above $2500 - 5min...

🛡️ RISK MANAGEMENT

Built-in protections:

  • Max position per market
  • Total capital limits
  • Strategy diversification
  • Event resolution monitoring
  • Inventory risk tracking

Circuit breakers:

Daily loss > 5% → Pause 1 hour
Total drawdown > 10% → Alert + reduce sizes
API rate limits → Auto-throttle
Failed trades > 10 → Pause strategy

🎯 PERFORMANCE TARGETS

Conservative (Parity + Tail-End):

Capital: $10,000
Monthly ROI: 8-15%
Monthly profit: $800-$1,500
Trades/day: 10-30
Win rate: 92-98%

Aggressive (All Strategies):

Capital: $50,000
Monthly ROI: 15-30%
Monthly profit: $7,500-$15,000
Trades/day: 50-200
Win rate: 80-90%

Real stats (from research):

  • $40M extracted April 2024 - April 2025
  • Top users: $1K-$2.2M profits
  • Bot win rates: 70-99%
  • Opportunities duration: 2s-48h

📊 API USAGE

Gamma API (Market Discovery):

GET /markets - Get all active markets
GET /markets/{id} - Get market details
Rate limit: 60 req/min

CLOB API (Trading):

GET /book?token_id={id} - Order book
GET /midpoint?token_id={id} - Midpoint price
GET /price?token_id={id}&side={BUY/SELL} - Best price
POST /order - Place order
DELETE /order - Cancel order

Rate limit: Public 60/min, Authenticated 100/min

WebSocket (Real-time):

wss://ws-subscriptions-clob.polymarket.com/ws/market/{token_id}
- Order book updates
- Trade updates
- Price changes

🔒 SECURITY & TRANSPARENCY

What it does:

  • ✅ Scans public Polymarket markets
  • ✅ Places orders via CLOB API
  • ✅ Uses your Polygon wallet
  • ✅ Trades with your USDC
  • ✅ Reports via Telegram

What it requires:

  • ✅ Polymarket API credentials
  • ✅ Polygon wallet private key
  • ✅ USDC balance on Polygon
  • ✅ Telegram bot (alerts)

Safeguards:

  • ✅ Position size limits
  • ✅ Strategy diversification
  • ✅ Event resolution monitoring
  • ✅ Circuit breakers
  • ✅ Full audit trail

Never:

  • ❌ Trades without detection
  • ❌ Exceeds position limits
  • ❌ Bypasses risk controls

⚠️ IMPORTANT NOTES

Polymarket specifics:

  • Operates on Polygon (low gas fees)
  • Trades settle on-chain (USDC)
  • Events resolve when outcome known
  • Markets can be cancelled/invalid
  • Fees: 2% on winners (some markets)

Capital requirements:

  • Minimum: $1,000 (limited opportunities)
  • Recommended: $5,000-$10,000
  • Professional: $50,000+

Legal:

  • Check local regulations
  • Prediction markets may be restricted
  • Use at your own risk
  • Not financial advice

Version: 1.0.0
License: MIT
Author: Georges Andronescu (Wesley Armando)

SCAN EVERYTHING. TRADE EVERYTHING. PROFIT EVERYWHERE. 🎯💰


END OF SKILL

Files

5 total
Select a file
Select a file to preview.

Comments

Loading comments…