Add oracle3 to Example Projects

[YichengYang-Ethan]

Adds oracle3 — an open-source autonomous trading agent for prediction markets — to the Agent Frameworks & Examples > Example Projects section, alongside polymarket-marketmaking, nevbot, and kalshi-price-monitor.

Why this fits

oracle3 directly matches the list's stated scope ("APIs, datasets, tools, and research for building on prediction markets — for developers and AI agents, not dashboards"):

• Multi-venue execution: trades Kalshi, Polymarket, and Solana DFlow from one strategy file
• Wang Transform pricing engine: closed-form risk-neutral measure calibrated by hierarchical maximum-likelihood on 291,309 resolved binary contracts — a quantitative approach to fair-value beyond LMSR/CLOB midpoint
• Eight constraint-based arbitrage strategies: cross-venue, complementary, and implication arbitrage with Kelly-sized execution and dual-layer risk management
• On-chain audit trails: Jito bundles + Solana Memo program for every trade

Quality signals

• 178 stars, Apache 2.0 license, v1.1.1
• 633 tests, mypy/ruff clean
• Zenodo DOI: 10.5281/zenodo.20062549 (versioned, archived in Software Heritage)
• Companion working paper "Pricing Prediction Markets: Risk Premiums, Incomplete Markets, and a Decomposition Framework" (SSRN 6468338)
• JOSS submission pending

Diff

Single-line addition to README.md, matching the existing entry format.

[YichengYang-Ethan]

Thanks for maintaining this list — the developer-focused framing (APIs/datasets/tools rather than dashboards) is exactly the right signal-to-noise filter for prediction markets. Happy to adjust the entry length or move it to a different section if you'd prefer (CLI Tools or Strategy & Market Structure could also work). The Wang Transform calibration data and eight arbitrage strategies are documented in the SSRN paper if anyone is curious about the methodology.