← Top 100 Full directory (1,587)

StratProof

Dev Tools Remote ★ 1,118 6 tools

StratProof runs honest backtests: real fees, real slippage, walk-forward validation. Six public tools, no OAuth. Call prove_strategy to test any strategy description ("Buy ETH when RSI(7) below 25, exit on 1.5% gain") and get a CONFIRMED / MIXED / DEBUNKED verdict with diagnosis. Call get_live_proof to see how our adaptive engine is actually performing in paper trading right now. Data sources: Binance spot (3 years), aggregated research output from 7,000+ tested candidates.

Connect

MCP endpoint

https://stratproof--pmort2222.run.tools

How to connect

Smithery (hosted)

1. Open https://smithery.ai/servers/pmort2222/stratproof
2. Click Connect and complete OAuth in your MCP client (Claude, Cursor, VS Code, etc.)
3. MCP endpoint: https://stratproof--pmort2222.run.tools

Tools (6)

prove_strategy
Run StratProof's honest backtest on a trading strategy. Tests the strategy description against 3 years of real Binance data with real fees and slippage, returns a CONFIRMED / MIXED / DEBUNKED verdict plus detailed statistics. This is the same engine that powers the public /prove page — no simplification, no shortcuts. Use this whenever a user describes a trading strategy and wants to know whether it would actually make money with real fees.
get_live_proof
Get StratProof's current live paper-trading statistics. Returns v1_baseline (fixed 22-strategy pool, the "control group") and v2_adaptive (self-building research engine, the "experimental group") side-by-side: trade count, win rate, average P&L, cumulative P&L. This is the transparency moat — StratProof publishes every trade win AND loss publicly. Use this when users ask about StratProof's credibility, track record, or "does your thing actually work."
get_research_findings
Get StratProof's research-lab findings — aggregated output from 7,000+ strategy candidates tested by the adaptive research engine. Returns the list of graduated strategies (those that passed all 5 validation gates) with their backtest stats (avg P&L, win rate, Sharpe, deflated Sharpe fluke probability). StratProof publishes losers AND winners; this tool returns both categories. Use this when a user asks "what strategies actually work" or wants to compare their idea to what the research lab has a
get_regime_insights
Get StratProof's regime-classifier validation results. Every night, StratProof runs a Welch's t-test on every strategy, per regime pair, to check whether its regime labels (ranging_low_vol, trending_high_vol, etc.) actually separate different outcome distributions. Returns aggregate significance stats, per-timeframe breakdown (1-hour / 4-hour / daily), top regime specialists, and a time-series history. If the classifier is random noise, <5% of pair-tests pass p<0.05; observing 20%+ means the lab
get_shootout_results
Get StratProof's strategy shootout — 20 popular "YouTube-famous" trading strategies tested with ZERO-FEE assumptions (what most backtesters show) versus REAL fees on Binance, Kraken, and Coinbase. Demonstrates the fee-drag gap: most strategies that look profitable at 0% fees lose money on every real exchange. Use this when a user asks about why backtests lie, about fee drag, about whether popular strategies actually work, or wants concrete evidence that honest backtesting matters.
get_stats
Get StratProof site-wide stats — total trades across all engines, unique coins tracked, backtests run, and metadata. A quick summary for "tell me about StratProof" questions or for verifying the site is live and producing data.