Think MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 23 installs a week from npm.
Intent security pre-flight checks for autonomous AI agents.
Today is the operative word: we check Think MCP 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 think-mcp -- npx -y @agentutil/think-mcp
{
"mcpServers": {
"think-mcp": {
"args": [
"-y",
"@agentutil/think-mcp"
],
"command": "npx"
}
}
}
[mcp_servers.think-mcp]
command = "npx"
args = ["-y", "@agentutil/think-mcp"]
{
"mcpServers": {
"think-mcp": {
"args": [
"-y",
"@agentutil/think-mcp"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"think-mcp": {
"args": [
"-y",
"@agentutil/think-mcp"
],
"command": "npx"
}
}
}
Read-only smart-contract security intelligence for autonomous agents.
License check, outdated deps, security for AI agents
Security guardrails for AI agent payments
Technology intelligence feed for AI agents: tech stack, compliance, security.
Security audit for AI agents — scan code/diffs for leaked secrets, check deps via OSV.
Pre-deploy security auditor for AI agent code — the risks generic SAST misses.
Website intelligence for AI agents: tech stack, hosting, security posture, SEO, and DNS.
MCP security trust-check for agents before installing MCPs, Skills or tools.
Answers built from our own checks of this server.