Polymarket 48h Weather Distribution Trader

Trades mispricings in weather temperature-bin markets by reconstructing the implied probability distribution across bins for the same city and date, detectin...

Security Scan

The skill's code, instructions, and required credential (SIMMER_API_KEY) are consistent with a Polymarket temperature-bin arbitrage trader that defaults to paper trading; nothing in the files points to unrelated credential access or surprising behavior, but the package origin is unknown so exercise caution before enabling live trading.

Purpose & Capability

The skill's name and description (temperature-bin distribution arbitrage on Polymarket) align with the actual code and instructions: it searches markets, reconstructs distributions, detects sum/monotonicity violations, and trades via the Simmer SDK. Note: the top-level registry summary in the prompt claimed 'Required env vars: none' and 'Primary credential: none', but both clawhub.json and SKILL.md declare SIMMER_API_KEY as required — this metadata mismatch is an inconsistency to be aware of but does not imply maliciousness.

Instruction Scope

SKILL.md and trader.py limit actions to market discovery, parsing, distribution construction, ranking violations, and placing trades through the SimmerClient. The instructions and code only read SIMMER_* tunables and the SIMMER_API_KEY; they do not instruct reading arbitrary local files, system credentials, or sending data to unexpected external endpoints beyond the declared Simmer/Polymarket pathways.

Install Mechanism

There is no explicit install script, but clawhub.json declares a pip dependency on 'simmer-sdk' (PyPI/GitHub URLs provided in SKILL.md). Relying on a PyPI package is reasonable for a trading client but introduces typical supply-chain risk (a malicious or compromised PyPI package could be harmful). The skill does not download arbitrary archives or use unknown shorteners.

Credentials

The only high-value credential requested is SIMMER_API_KEY (trading authority), which is proportional to a trading skill. The other SIMMER_* environment variables are tunables for risk limits and are reasonable. The code does not request unrelated credentials (AWS keys, GitHub tokens, etc.).

Persistence & Privilege

The skill is not forced to always run (always:false), autostart is false, and cron is null — it will not run automatically by default. The skill can be invoked autonomously by the agent (platform default), which is expected for skills that can trade; there is no evidence it modifies other skills or system-wide settings.

Assessment

This skill appears to do what it says: reconstruct distributions from Polymarket temperature-bin markets and trade via the Simmer SDK. Before enabling live trading: (1) Keep the SIMMER_API_KEY secret and give it the minimum permissions possible; treat it as a high-value credential. (2) Test extensively in paper mode (default) and verify behavior and sizing limits. (3) Inspect or vet the 'simmer-sdk' package (check the PyPI/GitHub repository, recent commits, and package maintainers) because pip dependencies are a common supply-chain risk. (4) Note the registry metadata mismatch: the prompt's summary said no required env vars, but the code requires SIMMER_API_KEY — ensure the platform will not inject unexpected credentials. (5) If you plan to run unattended automation, audit the code path that executes live trades (search for the live flag path in trader.py) and consider using a scoped API key or withdrawal limits on the exchange. If you want higher assurance, request the full untruncated trader.py for an independent review of the live-trade execution logic and any network calls.

48h Weather Distribution Trader

This is a template. The default signal is distribution-sum and monotonicity violation detection across weather temperature-bin markets — remix it with weather API feeds, ensemble forecasts, or cross-city correlation models. The skill handles all the plumbing (market discovery, distribution construction, trade execution, safeguards). Your agent provides the alpha.

Strategy Overview

Polymarket lists multiple temperature bins for the same city and date:

• "Will the highest temperature in Munich be 8C on March 28?" = 40%
• "Will the highest temperature in Munich be 9C on March 28?" = 45%
• "Will the highest temperature in Munich be 10C on March 28?" = 16%

These bins form a probability distribution that must sum to ~100%. When they don't, individual bins are mispriced. Additionally, cumulative markets ("X C or below", "X C or higher") impose monotonicity constraints.

The Edge: Distribution Arbitrage for Temperature Markets

In traditional markets, discrete outcome probabilities must sum to 1.0 — this is a fundamental axiom. On Polymarket, each temperature bin trades independently with its own order book and liquidity. Retail treats each bin as an isolated bet without checking the full distribution.

Violation Type 1: Sum Deviation

All exact bins for a (city, date) must sum to ~100%:

P(8C) + P(9C) + P(10C) + P(11C) + ... = 100%

If the sum is 108%, at least one bin is overpriced. If the sum is 92%, at least one bin is underpriced.

Violation Type 2: Cumulative Monotonicity Break

Cumulative markets must be monotonic:

P(<=8C) <= P(<=9C) <= P(<=10C)    [or_below: increasing]
P(>=10C) >= P(>=9C) >= P(>=8C)    [or_higher: decreasing with temp]

If a lower threshold has a higher "or below" probability than a higher threshold, the curve is broken.

Why This Works

1. Retail trades in silos — most users view each temperature bin independently and don't cross-reference the full distribution
2. No market maker enforcing consistency — unlike bookmakers who balance their book, Polymarket has no mechanism to keep bins summing to 100%
3. Mathematical, not opinion — the violations are provable inconsistencies in the probability axioms
4. Many cities, daily resolution — 14 cities with daily temperature markets create a large opportunity surface

Signal Logic

1. Discover all weather temperature markets via keyword search
2. Parse each question: extract city, temperature value, date, and type (exact/or_below/or_higher)
3. Group into distributions by (city, date)
4. For each distribution with 2+ bins:
   • Check if exact bins sum to ~100% (tolerance configurable via SIMMER_SUM_TOLERANCE)
   • If sum > 105%: identify and sell the most overpriced bin (highest relative to neighbors)
   • If sum < 95%: identify and buy the most underpriced bin (lowest relative to neighbors)
   • Check monotonicity on cumulative bins
5. Rank violations by magnitude
6. Trade only violations that also pass threshold gates (YES_THRESHOLD / NO_THRESHOLD)
7. Size by conviction (violation magnitude + threshold distance), not flat amount

Supported Cities

Chengdu, Shenzhen, Munich, Dallas, Austin, San Francisco, Seoul, Chicago, Wuhan, Miami, Seattle, Los Angeles, Denver, New York City.

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_SUM_TOLERANCE	0.05	Allowed deviation from 100% sum before trading

Edge Thesis

Weather temperature markets on Polymarket are structured as discrete probability distributions. Each bin trades independently, but they are mathematically constrained to sum to 100%. When retail order flow pushes individual bins without propagating to the full distribution, the sum deviates — creating pure mathematical arbitrage. This skill reconstructs the distribution, finds where the axioms break, and trades the repair.

Dependency

simmer-sdk by Simmer Markets (SpartanLabsXyz)

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