"Bootstrap" matching MCP servers:

Local-first backtesting engine with built-in overfitting detection (PBO, deflated Sharpe, bootstrap CI, walk-forward) and a native MCP server for AI agents to validate trading strategies.

Open-source MCP server that exposes Cegid APIs (accounting, third parties, documents, HR) as tools for LLM agents via the Model Context Protocol. Currently in early bootstrap phase with core structure in place.

A local, read-only MCP server (plus CLI) over your own stock/ETF transaction log. A deterministic core computes drawdown-first risk with bootstrap confidence intervals and validates targets with walk-forward verdicts. The assistant narrates, but the number fence makes it structurally impossible for the model to produce a figure.

Enables statistical validation of whether a trading backtest’s edge is real using bootstrap-resampling and reshuffling Monte Carlo methods, including confidence intervals, drawdown-path percentiles, expected value calculations, and prop-firm challenge pass-probability simulation. It also compares multiple win-rate/risk-reward geometries by simulated pass rate.