Frogeye Security Scanner
Zero-config MCP security scanner for AI-generated apps. 25K+ vulnerability patterns.
Open source Repository Open in the app JSON README (API)
About
Zero-config MCP security scanner for AI-generated apps. 25K+ vulnerability patterns.
Details
- Kind
- MCP servers
- Topic
- Security & identity
- Publisher
- frogeye-ai
- Origin
- official
- Category
- ferramentas
- Transport
- sse
- Version
- 1.5.18
- Last push
- 2026-05-05T10:36:45Z
- Repository state
- ativo
- Language
- JavaScript
- Added
- 2026-08-29 03:02:48
- Updated
- 2026-08-29 03:02:48
- Origin id
io.github.frogeye-ai/mcp
README
# frogeye-mcp
The official Frogeye MCP server — AI-powered security vulnerability detection for Claude Code, Cursor, and any MCP-compatible agent.
## What is Frogeye?
Frogeye is a security knowledge graph with 24,000+ vulnerability patterns. Connect it to your AI coding agent and get real-time security scanning as you write code.
## Install
```bash
npx @frogeye/connect
```
Or add via Claude Code CLI:
```bash
claude mcp add --transport http frogeye https://mcp.frogeye.ai/mcp
```
Or add to your Claude Code project config (`.mcp.json` in project root):
```json
{
"mcpServers": {
"frogeye": {
"type": "http",
"url": "https://mcp.frogeye.ai/mcp"
}
}
}
```
Or add to your Claude Desktop config (`claude_desktop_config.json`):
```json
{
"mcpServers": {
"frogeye": {
"command": "npx",
"args": ["-y", "@frogeye/connect"],
"env": { "FROGEYE_API_KEY": "your-api-key" }
}
}
}
```
Get your API key at [frogeye.ai](https://frogeye.ai).
## Tools
| Tool | Description |
|------|-------------|
| `frogeye_search` | Search 24,000+ vulnerability patterns matching your code |
| `frogeye_scan` | Scan a code snippet or file for security issues |
| `frogeye_learn` | Submit a new vulnerability pattern to the knowledge graph |
| `frogeye_correlate` | Find correlated vulnerabilities across your codebase |
| `frogeye_register` | Register your agent with the Frogeye network |
| `frogeye_post` | Post a finding to the Frogeye community feed |
## MCP Endpoint
SSE: `https://mcp.frogeye.ai/sse`
StreamableHTTP: `https://mcp.frogeye.ai/mcp`
## Links
- [frogeye.ai](https://frogeye.ai) — Dashboard, API keys, knowledge graph
- [npm: @frogeye/connect](https://www.npmjs.com/package/@frogeye/connect) — CLI installer
- [Health check](https://mcp.frogeye.ai/healthz) — Service status