# Varrd

Varrd — an MCP server that turns natural language trading ideas into tested quantitative strategies. It checks market data and computes exact entry, stop and profit levels.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 2427 installs
- Author: augiemazza
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/augiemazza/varrd
- Updated: 2026-03-01
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/varrd

## What it does

The server links your coding agent to a quantitative research engine. When you describe a setup, `varrd_ai` evaluates historical market records, tests the pattern, and calculates Sharpe ratios along with exact entry and exit price marks. It enforces safeguards against lookahead bias and sample contamination directly in the pipeline.

Beyond testing user ideas, `autonomous_varrd_ai` searches for novel patterns on its own, while `get_briefed` pulls current market news relevant to your saved rules. You can locate and review stored strategies using `search` and `get_hypothesis`.

Deep AI research calls consume credits, requiring an upfront purchase of at least $5 via Stripe or Base USDC. Running simple checks on active setups and inspecting your balance remain free.

## Requirements
- Remote server: `https://varrd--augiemazza.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http varrd https://varrd--augiemazza.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add varrd --url https://varrd--augiemazza.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "varrd": {
            "url": "https://varrd--augiemazza.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "varrd": {
            "type": "http",
            "url": "https://varrd--augiemazza.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http varrd https://varrd--augiemazza.run.tools
```

## When to use it
- test trading ideas against historical market datasets
- compute exact entry and exit price levels
- discover quantitative setups without manual charting
- generate a news digest tied to your active trade setups

## Example prompts
- Test an opening range breakout pattern on 1-hour candles and show the trade levels.
- Search my saved hypotheses for setups with the highest Sharpe ratio.
- Prepare a market news briefing focused on my current trading rules.

## Similar
- [VARRD — Statistically Validated Trading Edges + AI Research Engine](https://affpapa.org/en/nevibecode/mcp/varrd-statistically-validated-trading-edges-ai-research-engine.md) — MCP server · ★ 25 repo stars
- [longbridge-quant](https://affpapa.org/en/nevibecode/skills/longbridge-quant.md) — skill · 4300 installs
- [quantitative-research](https://affpapa.org/en/nevibecode/skills/quantitative-research.md) — skill · 2700 installs
- [StratEvo](https://affpapa.org/en/nevibecode/mcp/stratevo.md) — MCP server · ★ 34 repo stars
- [QuantContext](https://affpapa.org/en/nevibecode/mcp/quantcontext.md) — MCP server · ★ 10 repo stars
- [pine-backtester](https://affpapa.org/en/nevibecode/skills/pine-backtester.md) — skill · 468 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
