ClawForge and finclaw
Both projects are open-source quantitative finance engines for AI trading agents, making them competitors, as both aim to provide a comprehensive platform for backtesting, strategy evolution, and live trading with AI integration.
About ClawForge
leionion/ClawForge
OpenClaw AI trading agents - Skill Forge, Chat, BankrBot, Polyclaw, Alpaca, Kalshi, Whale TrackingOpenClaw AI trading agents OpenClaw AI trading agents
Demand decomposition via keyword + LLM (Chutes) breaks natural language trading requests into atomic skills, which execute locally or route through the OpenClaw gateway with automatic fallback to direct LLM when the gateway is busy. Integrates across six trading venues (Alpaca stocks, BankrBot crypto, Polymarket/Kalshi prediction markets, Whale Tracking on-chain), plus portfolio management, backtesting, and DCA automation—all unified in a Streamlit dashboard. Built on Python with OpenClaw as the agent runtime and Groq/Chutes as the language backbone, enabling both standalone skill execution and multi-agent orchestration.
About finclaw
NeuZhou/finclaw
AI-native quantitative finance engine. Quotes, backtesting, paper trading, strategy evolution, and MCP server.
Leverages genetic algorithms to autonomously evolve trading strategies across 484 technical, sentiment, and DRL factors—eliminating manual signal design—validated via walk-forward backtesting and Monte Carlo simulation. Supports crypto, A-shares, and US markets with zero API key setup, exposes strategies as MCP tools for Claude/Cursor integration, and employs NSGA-III multi-objective optimization to discover Pareto-optimal strategy frontiers rather than single fitness scores.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work