MCP Shield runs on your own machine — the client starts it, so there is no endpoint to ping. 34 installs a week from npm. Last commit 9 Jun 2026.
Security scanner for MCP servers. SSRF, path traversal, injection, auth, secrets. Grade A-F.
Today is the operative word: we check MCP Shield every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add mcp-shield -- npx -y @sophiacave/mcp-shield
{
"mcpServers": {
"mcp-shield": {
"command": "npx",
"args": [
"-y",
"@sophiacave/mcp-shield"
]
}
}
}
[mcp_servers.mcp-shield]
command = "npx"
args = ["-y", "@sophiacave/mcp-shield"]
{
"mcpServers": {
"mcp-shield": {
"command": "npx",
"args": [
"-y",
"@sophiacave/mcp-shield"
]
}
}
}
{
"servers": {
"mcp-shield": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@sophiacave/mcp-shield"
]
}
}
}
Security scanner for MCP servers and skills: Unicode injection, patterns, secrets.
Security scanner for MCP servers - detects tool poisoning and injection
Scan for prompt injection, secrets, PII, and vet MCP servers before installation
Security scanner for AI agents: prompt injection, leaked secrets, PII, and SSRF-risk URL detection.
Scans MCP servers for tool poisoning, prompt injection and supply chain risks.
MCP server security scanner: detects prompt injection, credential leaks, SSRF, tool poisoning.
Prompt-injection shield for agents: read pages, scan content, audit MCP servers and skills. x402.
Multi-layer security scanner for MCP servers and agent skills (injection, exfiltration)
Answers built from our own checks of this server.