walkforward-validator-mcp

An MCP server for walk-forward analysis of trading strategies — Walk-Forward Efficiency ratio, parameter-stability scoring, lock-after-optimization audits, and WFO window generation.

The problem this solves

A strategy optimized on one historical window and never re-validated on a fresh, unseen window is a curve fit until proven otherwise. Walk-forward analysis is the standard fix, but building the rolling windows correctly and scoring whether a parameter surface is a robust plateau or a fragile spike is easy to get subtly wrong by hand.

Tools

walk_forward_efficiency

Computes the Walk-Forward Efficiency ratio — out-of-sample performance as a fraction of in-sample performance — the core signal for whether an optimization generalizes.

parameter_stability_score

Scores a parameter surface for fragile curve-fit spikes vs. robust plateaus, flagging optimizations that only work at one exact parameter value.

lock_after_wfo_check

Audits whether parameters were genuinely locked after the walk-forward optimization step, or quietly re-tuned against the "out-of-sample" data — the mistake that silently invalidates a WFO result.

walk_forward_window_generator

Generates correctly non-overlapping rolling in-sample/out-of-sample windows for a given date range and step size.

Use it

Hosted (recommended): MCPize — free tier, paid Pro tier for higher limits.

Self-host:

npm install
node server.js

Part of the AlgoForge suite

Prop-firm and quant-validation tools for algo traders: prop-rules-mcp, trade-journal-mcp, payout-calc-mcp, econ-calendar-mcp, montecarlo-validator-mcp, overfitting-audit-mcp, pinescript-audit-mcp, backtest-cost-sensitivity-mcp, pinescript-mcp.

License

MIT