Cross-Platform Arbitrage Surge Dominates Polymarket Dev Activity

Cross-Platform Arbitrage Surge Dominates Polymarket Dev Activity

**Date: 2026-06-26**

Lead Story: Kalshi-Polymarket Arbitrage Is Now a Crowded Trade

Three separate repositories published or updated today target the same fundamental opportunity: price discrepancies between Polymarket and Kalshi on correlated or identical markets. `pranavrao87/kalshi_poly_engine` (created today, 0 stars) and `sftgroup/predx` (also created today) both describe real-time scraping engines built to surface cross-platform mispricings. Neither shows meaningful stars or forks yet, but the timing is notable — two independent developers shipping similar infrastructure on the same day suggests the opportunity is well-known and competition for the edge is intensifying.

`matthewnyc2/arbitrage` takes a more defensible angle: it focuses specifically on NegRisk multi-outcome arbitrage *within* Polymarket, where the sum of outcome probabilities across a single event can violate the no-arbitrage bound. The repo claims 60 passing tests and a paper-trading mode, which is more credible production scaffolding than most entries in this space. If the NegRisk scanner is as robust as described, it represents a repeatable, lower-risk strategy that doesn't depend on Kalshi API stability or cross-chain settlement friction.

Infrastructure Watch: PMXT Crosses 1,900 Stars

`pmxt-dev/pmxt` continues to be the most consequential repository in the Polymarket developer ecosystem. At 1,917 stars and 228 forks, this CCXT-style unified API for Polymarket, Kalshi, and additional venues is now a meaningful piece of market infrastructure. Its existence lowers the barrier for any quant shop already running CCXT-based crypto strategies to port directly into prediction markets. Operators should treat this as a structural factor: as onboarding friction falls, expect more sophisticated capital to enter and compress obvious mispricings faster.

Strategy Signals

**Weather markets under pressure.** `BallesJr/polymarket-weather-edge` describes a Gaussian probability model using live METAR observations and Open-Meteo forecasts to fade overpriced NO positions in temperature markets. The methodology is sound — METAR data is high-frequency and publicly available, and temperature resolution markets have historically shown retail-driven mispricing. If this bot is paper-trading profitably, expect live deployment and tightening spreads in weather categories within weeks.

**Political cross-market arbitrage.** `nayrbryanGaming/arbiter` explicitly targets no-arbitrage violations between related election markets, sizing with fractional Kelly and requiring backtested out-of-sample edge before live execution. The stated discipline — paper-trading by default, survival over profit — is atypical for this space and suggests a more experienced developer. Political markets with correlated outcomes (e.g., state-level races implied by national markets) are a known hunting ground; this bot formalizes the scan.

**Rust execution engine.** `Degenapetrader/EVPOLY` (45 stars, 14 forks) is one of the few Polymarket execution engines written in Rust, with multiple strategy loops and a shared risk arbiter. The source-available, non-commercial license limits direct forks for production use, but the architecture is worth studying. Low-latency execution is increasingly relevant as more bots compete on the same signals.

Security Alert

`Rolymarket/Polymarket-Arbitrage-Trading-Bot-Python` (258 stars, 230 forks) warrants caution. The description is pure keyword stuffing — the phrase "polymarket arbitrage bot" and "polymarket trading bot" repeat roughly a dozen times with no technical detail. A 35KB repository with that star-to-content ratio is a red flag. Do not connect API keys or wallet credentials to any code sourced from this repository without a full manual audit. Treat it as a likely credential harvester until proven otherwise.

Operator Takeaways

• Cross-platform Kalshi-Polymarket arbitrage is now a crowded, competitive trade. Entry without a latency or information advantage is increasingly unattractive.
• NegRisk within-Polymarket violations remain more defensible for smaller operators; `matthewnyc2/arbitrage` is worth monitoring for methodology.
• PMXT's growth signals accelerating institutional interest. Adjust expectations for mispricing duration accordingly.
• Avoid any bot repository with keyword-stuffed descriptions and disproportionate stars relative to code size.