stillos-edge-gate
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@stillos-edge-gateanalyze this trade log for a real edge"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
stillos-edge-gate
A callable trust primitive. Grade a track record before you commit capital.
A real edge is not a prediction — it is a repeatable, out-of-sample, fee-adjusted decision advantage that survives a regime change. This tells you, in one word, whether a strategy has one:
REAL_EDGE | REGIME_LUCK | NEGATIVE_EV | INSUFFICIENT_DATAFail-closed: anything that isn't a surviving, fee-adjusted, regime-stable edge is rejected. Every verdict ships with a signed, independently verifiable receipt.
Install
npx stillos-edge-gate grade trades.jsonor add as an MCP server (Claude Desktop / any MCP client):
{ "mcpServers": { "stillos-edge-gate": { "command": "npx", "args": ["stillos-edge-gate", "mcp"] } } }Related MCP server: DJZS Trust MCP
Input
[ { "t": "2026-06-01T12:00:00Z", "price": 0.62, "side": "no", "outcome": 1 }, ... ]price = entry price 0..1 · side = yes|no · outcome = 1 won / 0 lost.
How it decides
chronological 70/30 out-of-sample holdout (most-recent 30% is the test)
fee + slippage adjusted per trade
significance test (t ≥ 1.5) on fee-adjusted EV
REAL_EDGEonly if +EV in both splits after fees, significant.
Why trust it
We ran it on our own live trading book first. It returned NEGATIVE_EV — and we published the signed receipt. We don't sell edges; we tell you whether yours is real.
Library
const { gradeStrategy } = require('stillos-edge-gate');
const v = gradeStrategy(trades); // { verdict, train, test, t_stat, reasons }This server cannot be deployed
Maintenance
Related MCP Connectors
Honest A-F grades for trading strategies, backtested on real out-of-sample data. No hype.
Backtest trading strategies written in plain English, on real market data, with graded results.
- mcpOAuthcom.market-graphs
Praxis: published trading strategies re-run and audited — verdicts, claimed vs measured stats.
Deterministic pre-execution audit for trading agents. PASS/WAIT/FAIL, reproducible verdict_hash.
Related MCP Servers
- AlicenseAqualityAmaintenanceMost trading signals are noise. AlphaAssay puts them on trial — deflated Sharpe, out-of-sample, leakage forensics — and returns signed pass/fail verdicts anyone can verify. Methodology audits, not investment advice.17Apache 2.0
- AlicenseNot gradedqualityBmaintenanceDeterministic pre-execution audit for trading agents. PASS/WAIT/FAIL, reproducible verdict_hash.53MIT
- AlicenseBqualityAmaintenanceVerify a number before an agent asserts it — a Deflated Sharpe Ratio for backtest, plus eval-gap, subset-win, and judge-bias checks, with signed receipts anyone can verify offline.34MIT
- AlicenseNot gradedqualityBmaintenanceChecks whether a trading backtest survives its own statistics: deflated Sharpe, multiple-testing correction against a best-of-N-noise benchmark, minimum track record length, and fill realism. Takes no market data and no API keys, and cannot recommend a trade — it only reports that a result is weaker than claimed or not yet provable.MIT