Polymarket Bundle Cross Coin 5min Trader

Trades cross-coin divergence in 5-minute crypto Up/Down bundles on Polymarket. BTC, ETH, SOL, and XRP move together in practice -- when one coin diverges fro...

Security Scan

The skill's code, declared dependencies, and runtime instructions are coherent with its stated purpose (trading Polymarket 5‑minute cross‑coin divergences); required credentials and tunables align with trading behavior and the skill defaults to paper mode.

Purpose & Capability

Name/description claim Polymarket trading. The package depends on simmer-sdk and the only required secret is SIMMER_API_KEY, which is the trading authority the skill needs. The code parses Polymarket questions, groups 5‑min windows, computes deviations, and sizes trades — all consistent with the stated strategy.

Instruction Scope

SKILL.md and trader.py limit actions to market discovery, signal computation, safeguards, and trade execution. The skill reads only the declared environment tunables and the SIMMER_API_KEY. There are no instructions to read arbitrary local files or send data to unrelated external endpoints.

Install Mechanism

No install script included in the bundle, but clawhub.json lists a pip dependency on simmer-sdk (PyPI/GitHub links provided). Requiring a PyPI package is expected for this functionality but carries the usual moderate risk of third‑party package code — verify the simmer-sdk package and its versions before installation.

Credentials

Only SIMMER_API_KEY is required (declared). The script also reads multiple tuning env vars, all declared in clawhub.json. No unrelated credentials or system secrets are requested.

Persistence & Privilege

autostart is false and always is not set, so the skill is not forced into every agent run. The automaton entrypoint is declared (managed:true, entrypoint: trader.py) which lets the platform run it as an automaton if enabled — this is expected for a trading skill but you should keep autostart/cron disabled until you review behavior.

Assessment

This skill appears internally consistent and implements the described Polymarket divergence strategy. Before installing: 1) Review the simmer-sdk PyPI/GitHub package and its maintainers (supply-chain risk). 2) Keep autostart/cron disabled and test extensively in paper mode (default) until you are confident. 3) Only supply a SIMMER_API_KEY that has the minimum permissions you are willing to grant; consider using a limited/test account for live trades and rotate keys after testing. 4) Inspect the full trader.py (and any updates) for changes before switching to --live, and monitor trading activity if you enable live mode.

Bundle -- Cross-Coin 5min Divergence Trader

This is a template. The default signal is cross-coin divergence detection within 5-minute crypto Up/Down bundles -- remix it with on-chain correlation data or additional coins. The skill handles all the plumbing (market discovery, time window grouping, trade execution, safeguards). Your agent provides the alpha.

Strategy Overview

Polymarket lists "Bitcoin Up or Down", "Ethereum Up or Down", "Solana Up or Down", and "XRP Up or Down" markets for the same 5-minute windows throughout the day. Crypto assets are highly correlated -- when BTC moves up, ETH/SOL/XRP almost always follow. When one coin's market price diverges from the group consensus, it is structural mispricing that should revert.

Example: In the 10:05 AM ET window, if BTC=Up(60%), ETH=Up(55%), SOL=Down(40%), then SOL is the outlier. The group consensus is ~52% Up, but SOL is priced at 40% -- a 12% deviation that should close as correlated price action propagates.

Edge

Crypto correlation is the strongest cross-asset relationship in financial markets. BTC, ETH, SOL, and XRP have 30-day rolling correlations consistently above 0.80. Within a 5-minute window, the correlation is even higher because macro moves dominate and idiosyncratic moves are negligible. Polymarket prices each coin's Up/Down market independently -- different order books, different liquidity pools, different retail participants. When one coin's order book gets hit by a directional flow that doesn't appear in the others, the resulting price divergence is a mean-reversion opportunity.

Signal Logic

1. Discover active crypto Up/Down 5-minute markets via keyword search + get_markets(limit=200) fallback
2. Parse each question: extract coin (BTC/ETH/SOL/XRP), date, and time window
3. Normalize time windows to 5-minute buckets for cross-coin matching
4. Group markets by (date, time_bucket) -- find windows where 2+ coins have markets
5. For each window: compute group consensus (mean probability across all coins in the window)
6. Find outliers: any coin deviating > SIMMER_MIN_DEVIATION (default 10%) from group mean
7. Trade outlier toward group mean: buy YES if below, sell NO if above
8. Size by conviction (deviation magnitude), not flat amount

Remix Signal Ideas

• On-chain correlation feed: Pull real-time BTC/ETH/SOL/XRP price feeds from CoinGecko or Binance API -- compute rolling 5-min correlation and weight the mean-reversion trade by current correlation strength
• Volume-weighted consensus: Instead of simple mean, weight each coin's contribution to the group consensus by its Polymarket volume -- higher-volume coins anchor the consensus more strongly
• Momentum filter: Check if the outlier coin has idiosyncratic news (e.g., SOL network outage) that justifies decoupling -- skip those divergences
• Multi-window confirmation: Require divergence to persist across 2+ consecutive 5-min windows before trading -- filters out noise from order book mechanics

Safety & Execution Mode

The skill defaults to paper trading (venue="sim"). Real trades only with --live flag.

Scenario	Mode	Financial risk
python trader.py	Paper (sim)	None
Cron / automaton	Paper (sim)	None
python trader.py --live	Live (polymarket)	Real USDC

autostart: false and cron: null mean nothing runs automatically until configured in Simmer UI.

Required Credentials

Variable	Required	Notes
SIMMER_API_KEY	Yes	Trading authority. Treat as a high-value credential.

Tunables (Risk Parameters)

All declared as tunables in clawhub.json and adjustable from the Simmer UI.

Variable	Default	Purpose
SIMMER_MAX_POSITION	40	Max USDC per trade at full conviction
SIMMER_MIN_TRADE	5	Floor for any trade
SIMMER_MIN_VOLUME	5000	Min market volume filter (USD)
SIMMER_MAX_SPREAD	0.08	Max bid-ask spread
SIMMER_MIN_DAYS	0	Min days until resolution (0 = allow same-day)
SIMMER_MAX_POSITIONS	8	Max concurrent open positions
SIMMER_YES_THRESHOLD	0.38	Buy YES only if market probability <= this
SIMMER_NO_THRESHOLD	0.62	Sell NO only if market probability >= this
SIMMER_MIN_DEVIATION	0.10	Min deviation from group mean to trigger a trade

Edge Thesis

Traditional crypto exchanges maintain cross-asset consistency through arbitrage bots that trade correlated pairs in milliseconds. Polymarket has no such mechanism -- each coin's Up/Down market is an independent order book with its own retail participants. 5-minute crypto Up/Down markets are particularly vulnerable because:

• Short time windows mean correlation is near-maximum (macro dominates, idiosyncratic noise is minimal)
• Each coin attracts different retail communities (BTC maxis, ETH DeFi traders, SOL memecoin traders)
• Directional flow on one coin doesn't propagate to the others
• New 5-min windows open frequently, creating fresh opportunities every few minutes
• Retail prices each coin in isolation rather than as a correlated basket

This skill treats the cross-coin bundle as a correlated group and trades the outlier back toward the consensus.

Dependency

simmer-sdk by Simmer Markets (SpartanLabsXyz)

• PyPI: https://pypi.org/project/simmer-sdk/
• GitHub: https://github.com/SpartanLabsXyz/simmer-sdk