Scope Guard runs on your own machine — the client starts it, so there is no endpoint to ping. 41 installs a week from npm. Last commit 24 Apr 2026.
Hard scope boundaries to prevent AI agent drift
Today is the operative word: we check Scope Guard 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 scope-guard -- npx -y scope-guard-mcp
{
"mcpServers": {
"scope-guard": {
"args": [
"-y",
"scope-guard-mcp"
],
"command": "npx"
}
}
}
[mcp_servers.scope-guard]
command = "npx"
args = ["-y", "scope-guard-mcp"]
{
"mcpServers": {
"scope-guard": {
"args": [
"-y",
"scope-guard-mcp"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"scope-guard": {
"args": [
"-y",
"scope-guard-mcp"
],
"command": "npx"
}
}
}
Process management and monitoring for AI agents with strict security boundaries
Real estate data for AI agents: US parcel boundaries (tiles), owners, sale history, permits, hail.
Scoped deterministic SRD 5.2.1 rules checks for agents — applied rules cited, boundaries explicit.
AI agent relay — message business agents across company boundaries via A2A protocol
Provenance-aware memory format for AI agents: every claim carries its source, age and boundary.
Advanced filesystem operations with strict security boundaries for AI agents
Simulate hundreds of AI agents to predict how communities react to events and policies
Verify npm packages before your AI agent installs them: hallucinations, advisories, API drift.
Answers built from our own checks of this server.