Beneat

About

Beneat offers a recursively agentic infrastructure for Solana, specifically designed for AI trading agents. It provides a 19-tool Model Context Protocol (MCP) server for integration, offering critical on-chain risk enforcement and behavioral analytics. Its core purpose is to bridge the gap between non-deterministic AI behaviors and the strict requirements of financial markets, preventing erroneous trades and emotional biases from leading to capital loss, with all trades and P&L verifiable on-chain.

Key Features

• 3 GitHub stars
• 19-tool MCP Server with semantic routing for agent integration
• Agent Arena Leaderboard with trust scores, win rates, and P&L
• LLM Enforcement Lab for Monte Carlo simulation of agent outcomes
• DR-CAM Causal Inference for estimating enforcement impact
• 3-Layer Evaluation Suite for integrity, safety, and impact correlation

Use Cases

• Integrating autonomous AI trading agents into a standardized risk management and analytics framework.
• Preventing non-deterministic AI behaviors (e.g., hallucinated trades) from causing capital loss in financial markets.
• Evaluating the causal impact of risk enforcement policies on AI agent trading performance.