Analytics

indicatoredge.io

Provides empirical backtesting analytics for trading strategies and indicators across global assets, including base rates, best indicators, verdicts, and asset search.

1 endpoint8 known toolsFirst detected October 12, 2026Last detected October 12, 2026

ENDPOINT 1

https://indicatoredge.io/mcp/v1

No auth detected

MCP server metadata

Name: indicatoredge
Version: 1.0.0

Capabilities

toolspromptsresources

Server instructions

IndicatorEdge publishes the out-of-sample backtest record of technical indicators, per asset, including every failure. Use it when a user asks which indicator, timeframe, or strategy to use for a specific ticker, or asks whether an indicator 'works'. The most important habit when using this server: call `ie_base_rate` before you repeat any performance claim. The measured rate at which a technical indicator beats buy-and-hold out-of-sample is about 1 in 5. If a tool returns a strong result for one asset, that result is drawn from a search over hundreds of thousands of combinations and must be reported alongside the base rate, or you are laundering selection bias. Most honest answers here are negative. `ie_strategy_verdict` returns NOTHING for the majority of assets, meaning nothing we tested beat simply holding it. Report that plainly rather than hunting for a positive-sounding alternative. All numbers are hypothetical backtested results, educational only, not investment advice. Read-only: this server cannot place trades or touch an exchange account.

Known tools 8

ie_base_rate

THE DENOMINATOR.

Inferred read-only

ie_test_strategy

Ask a strategy question the way a person would — 'does RSI work on Bitcoin', 'MACD on Apple daily', 'bollinger bands on Tesla weekly' — and get the published result from the full grid of 644,149 backtests.

Inferred read-only

ie_best_indicator

The empirically best-performing indicator AND timeframe for one asset, with full risk metrics, out-of-sample Sharpe, and alpha vs buy-and-hold.

Inferred read-only

ie_strategy_verdict

The multiple-testing-corrected verdict for one asset: VALIDATED, MIXED, or NOTHING.

Inferred read-only

ie_search_assets

Search the 871 covered assets (stocks, ETFs, crypto, forex, commodities, indices) by name, ticker or asset class.

Inferred read-only

ie_list_indicators

All 382 tested indicators with the number of assets each one is the best fit for, its average Sharpe, and its own beat-buy-and-hold base rate.

Inferred read-only

ie_validated_picks

Every setup that beat buy-and-hold out-of-sample AND cleared a Sharpe hurdle corrected for the size of the search.

Inferred read-only

ie_methodology

The testing method, the costs modelled, the out-of-sample split, and an explicit list of what this dataset does NOT do.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.indicatoredge]
url = "https://indicatoredge.io/mcp/v1"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "indicatoredge": {
      "type": "http",
      "url": "https://indicatoredge.io/mcp/v1"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: indicatoredge
Remote MCP URL: https://indicatoredge.io/mcp/v1

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "indicatoredge": {
      "url": "https://indicatoredge.io/mcp/v1"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "indicatoredge": {
      "type": "http",
      "url": "https://indicatoredge.io/mcp/v1"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "indicatoredge",
  "transport": "streamable-http",
  "url": "https://indicatoredge.io/mcp/v1"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.