Power your agent
URL ของผู้เผยแพร่เอง ไม่ใช่เอนด์พอยต์ของ PlugMCP
Host a copy with our keys and logs.
https://neon_innovation_lab--mcp-security-auditor.apify.actor/mcpปุ่ม Claude ด้านบนจะเปิด Claude พร้อมกรอกชื่อและ URL ให้แล้ว คุณยังต้องยืนยันที่นั่นเอง
claude mcp add --transport http io.github.ansarii-mcp-security-auditor https://neon_innovation_lab--mcp-security-auditor.apify.actor/mcpRun in a terminal.
{ "mcpServers": { "io.github.ansarii-mcp-security-auditor": { "command": "npx", "args": [ "-y", "mcp-remote", "https://neon_innovation_lab--mcp-security-auditor.apify.actor/mcp" ] } }}claude_desktop_config.json
URL แบบรีโมตหรือคำสั่งบนเครื่องของคุณ? การติดตั้งสองแบบต่างกันอย่างไร
รันบน PlugMCP
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
เปิดA production-ready Model Context Protocol (MCP) Server providing AI coding assistants (Claude Desktop, Cursor, Windsurf, Cline) with native security analysis tools. It enables AI agents to statically audit other MCP servers, tool repositories, and local codebases for critical security vulnerabilities before onboarding or executing them.
This server implements the official Model Context Protocol specification (tools/list and tools/call):
audit_mcp_repositoryPerforms an automated, zero-execution static Abstract Syntax Tree (AST) security audit on a Git repository containing an MCP server implementation.
repository_url (string, required): Public Git URL of the MCP server repository to audit (e.g. https://github.com/example/mcp-server).sub_directory (string, optional): Specific subdirectory within the repository to inspect (useful for monorepos).subprocess.run, child_process.exec, shell invocation in tool execution blocks.os.path.realpath, boundary checks).trust_score (0–100): Weighted security index.grade ("A+", "A", "B", "C", "F"): Security letter grade.findings (array): Line-by-line vulnerability records with CWE identifiers and remediation instructions.markdown_report (string): Pre-formatted human-readable audit report.audit_local_directoryAudits a local filesystem directory containing MCP server source code before deployment.
directory_path (string, required): Absolute or relative filesystem path to the target directory.Add the server to your claude_desktop_config.json:
{
"mcpServers": {
"mcp-security-auditor": {
"command": "python3",
"args": ["/path/to/mcp-security-auditor/server.py"]
}
}
}
In Cursor, go to Settings → Features → MCP → Add New MCP Server:
mcp-security-auditorcommandpython3 /absolute/path/to/server.pyRun isolated via Docker using the pre-configured Dockerfile:
{
"mcpServers": {
"mcp-security-auditor": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/ansarii/mcp-security-auditor"
]
}
}
}
Once connected, your AI assistant can execute security pre-flight checks before installing tools:
User: "Audit the MCP server at https://github.com/modelcontextprotocol/servers before I install it."
Claude (Tool Call):
audit_mcp_repository({
"repository_url": "https://github.com/modelcontextprotocol/servers",
"sub_directory": "src/fetch"
})
Result:
{
"status": "success",
"target_name": "servers:src/fetch",
"trust_score": 95,
"grade": "A+",
"findings_count": 0,
"recommendation": "SAFE_TO_CONNECT"
}
Unlike runtime scanners that execute arbitrary stdio binaries (creating direct Remote Code Execution risks on the host), this MCP server uses pure static Abstract Syntax Tree (AST) analysis. It inspects Python, TypeScript, and JSON-RPC implementations without ever executing untrusted code.
MIT License — Copyright (c) 2026 Neon Innovation Lab.
Static AST security scanner detecting command injection, leaked secrets, and SSRF in MCP tools.
{ "mcpServers": { "io.github.ansarii-mcp-security-auditor": { "url": "https://neon_innovation_lab--mcp-security-auditor.apify.actor/mcp", "headers": {} } }}เครื่องมือ
ยังไม่มีรายการเครื่องมือ เซิร์ฟเวอร์แบบรีโมตจะได้รายการหลังการตรวจสอบสถานะครั้งแรก
FAQ
คัดลอกคอนฟิก http สำหรับ Cursor, Claude Desktop, Claude Code หรือ VS Code ได้ในหน้านี้ คอนฟิกนี้ใช้การติดตั้งของผู้เผยแพร่ ไม่ใช่พร็อกซีของ PlugMCP
ระเบียนนี้ไม่ได้ระบุตัวแปรสภาพแวดล้อมที่จำเป็น
MCP Security & Vulnerability Auditor เข้าถึงได้ผ่าน HTTP
ตรวจสอบล่าสุดเมื่อ 2026-11-08 Static AST security scanner detecting command injection, leaked secrets, and SSRF in MCP tools.