Four Arbitrage Bots Drop; Spain Blocks Polymarket

Four Cloneable Bots, Two Arb Races, and a Regulatory Shock

Today's scan surfaced eight new repositories, four of which are worth your time. The dominant theme is arbitrage — two separate teams have independently built cross-exchange scanners hitting overlapping platform sets, which either means there's a known inefficiency being hunted or both teams are about to learn how thin that edge really is. Either way, the code is instructive.

---

New Bots: What's Worth Cloning

**polymarketmaker** is the most structurally sophisticated drop today. It implements market making in discrete phases: spread capture, inventory management, and hard risk limits. This is the closest thing to a production-ready MM framework we've seen in open source Polymarket tooling. If you're building a maker strategy, start here and strip out what you don't need.

**godseye** is a signal generation layer that ingests live market data and routes outputs to live trading. The interesting part is architectural — it separates signal from execution cleanly, which makes it a reasonable base for anyone who wants to plug in their own alpha without rebuilding the execution stack.

**pred-arbitrage** scans for price divergence across three major prediction market platforms. It's functional but narrower in scope than the next entry.

**prediction-markets** hits four platforms in its cross-exchange arb scan. The overlap with pred-arbitrage is significant — both are hunting the same spread, likely with similar latency profiles. If you're deploying either, assume the edge is contested and model your fill rates pessimistically.

**poly-arb** is on monitor status. It has paper-trading simulation and a dashboard, which suggests it's earlier stage or built for research rather than live deployment. Worth revisiting in two weeks.

---

Security: Three Alerts, One That Matters

No confirmed malware in today's scan, but three alerts warrant attention.

A repository with a garbled username is showing structural hallmarks of a cloned malware template — obfuscated code, copied README, no commit history that makes sense. Do not run anything from it. Garbled or randomized usernames are a consistent pattern in credential-harvesting repos that front as trading tools.

A second garbled-username repo is presenting as a copy-trading panel. This one looks like a credential phishing front. Any tool that asks for your Polymarket API keys or wallet seed phrase through a UI you didn't build yourself should be treated as hostile.

Third: a Polymarket Telegram bot repo is showing star counts inconsistent with its age and activity. Likely star-farming to manufacture credibility. Don't weight GitHub stars as a trust signal for this category of tool.

---

Alpha Signals: Research Worth Reading

Three papers surfaced today with direct operational relevance.

The most immediately actionable: a pipeline that clusters prediction markets by topic using natural language, detecting implicit equivalences and price contradictions across fragmented listings. This is a systematic approach to the "same event, two markets, different prices" problem that anyone trading Polymarket manually has noticed. If you can implement even a lightweight version of this clustering, you have a structured arb scanner that doesn't depend on exchange integrations.

The merger arbitrage paper is worth reading for methodology rather than direct application. It uses long-context LLM reasoning over deal documents to forecast M&A outcomes. The context engineering techniques are transferable to any Polymarket contract with a large document trail — regulatory decisions, legal cases, technical reports.

The "financial epiplexity" paper is more theoretical but relevant if you're thinking about edge decay. It formalizes how much learnable structure a market contains given computational constraints. The practical implication: edge in prediction markets is real but bounded, and the bound tightens as more bots compete for the same signal.

---

What to Watch

Spain blocking Polymarket and Kalshi is the regulatory story of the week. Watch for copycat actions in other EU jurisdictions — this is the kind of move that travels. If you have exposure to European event markets or are routing through EU infrastructure, get clarity on your legal surface now.

The HN thread about threats against a trader over the Iran missile story hit 1,600 points. That's mainstream visibility for prediction market dysfunction. Expect more regulatory scrutiny of geopolitical contracts specifically.