KillerSkills runs on your own machine — the client starts it, so there is no endpoint to ping. 1 626 installs a week from npm. Last commit 31 Oct 2026.
Security verdicts for agent skills and MCP servers, from sandbox runs, before you install them.
Today is the operative word: we check KillerSkills 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 killerskills -- npx -y vet-skills
{
"mcpServers": {
"killerskills": {
"command": "npx",
"args": [
"-y",
"vet-skills"
]
}
}
}
[mcp_servers.killerskills]
command = "npx"
args = ["-y", "vet-skills"]
{
"mcpServers": {
"killerskills": {
"command": "npx",
"args": [
"-y",
"vet-skills"
]
}
}
}
{
"servers": {
"killerskills": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"vet-skills"
]
}
}
}
Check the trust/security score of MCP servers, agents, skills & CLIs before you install them.
Search and install 20,000+ security-graded MCP servers and agent skills.
MCP security trust-check for agents before installing MCPs, Skills or tools.
Security scanner for AI agent skills and MCP servers
Security grades (A-F) for MCP servers. Check one before you install or trust it. mcpgrade.dev
Scan for prompt injection, secrets, PII, and vet MCP servers before installation
Pre-install security scans for npm packages, MCP servers, and AI agents with cited verdict evidence.
Free, read-only security scanner for remote MCP servers, before you connect them.
Answers built from our own checks of this server.