# WaveGuard

WaveGuard checks metrics, text, logs, and time series for anomalies. It helps investigate suspicious records, token data, and wallet behavior.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 1318 installs
- Author: emergentphysicslab
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/emergentphysicslab/waveguard
- Updated: 2026-02-21
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/waveguard

## What it does

WaveGuard takes a small set of normal samples together with data under review. It then marks unusual items, gives a reason for each result, and reports a confidence level. Structured JSON rows, text, logs, and numeric arrays are supported. For time series, earlier windows establish the baseline while later windows are inspected.

The server exposes tools such as `waveguard_scan`, `waveguard_scan_timeseries`, `waveguard_fingerprint`, `waveguard_compare`, and `waveguard_health`. Other tools cover token risk, wallet behavior, trading volume, price manipulation, sequence drift, instability, and cascade risk. `waveguard_market_data` can retrieve crypto data from CoinGecko or DexScreener.

Authentication is not required for `waveguard_health`, so it provides a direct connection check. The endpoint is `https://waveguard--emergentphysicslab.run.tools`; setup instructions are available for Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and other listed clients.

## Requirements
- Remote server: `https://waveguard--emergentphysicslab.run.tools`

## Install

### Claude Code (native platform)

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

### Codex

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

### Cursor

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

### GitHub Copilot

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

### Gemini CLI

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

## When to use it
- spot outliers in JSON rows
- scan a numeric time series for anomalies
- assess a crypto token’s risk
- detect suspicious trading volume
- compare structural data fingerprints

## Example prompts
- Run `waveguard_scan` on these baseline and test rows, then explain every flagged anomaly.
- Use `waveguard_scan_timeseries` on this numeric array and inspect the latest windows for abnormal behavior.
- Fetch the token data with `waveguard_market_data`, then assess it with `waveguard_token_risk`.

## Similar
- [dt-obs-predictive-analytics](https://affpapa.org/en/nevibecode/skills/dt-obs-predictive-analytics.md) — skill · 2500 installs
- [Time Series Analysis](https://affpapa.org/en/nevibecode/skills/time-series-analysis.md) — skill · 617 installs
- [azure-ai-anomalydetector-java](https://affpapa.org/en/nevibecode/skills/azure-ai-anomalydetector-java.md) — skill · ★ 3232 repo stars
- [easy-predict](https://affpapa.org/en/nevibecode/mcp/easy-predict.md) — MCP server · 140 installs
- [kibana-anomaly-detection](https://affpapa.org/en/nevibecode/skills/kibana-anomaly-detection.md) — skill · 3500 installs
- [elasticsearch-anomaly-detection-explainer](https://affpapa.org/en/nevibecode/skills/elasticsearch-anomaly-detection-explainer.md) — skill · 1500 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
