Actenon Kernel runs on your own machine — the client starts it, so there is no endpoint to ping. 160 installs a week from pypi. Last commit 27 Jul 2026.
Verify consequential AI actions before they execute.
Today is the operative word: we check Actenon Kernel 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 kernel -- uvx actenon-kernel
{
"mcpServers": {
"kernel": {
"args": [
"actenon-kernel"
],
"command": "uvx"
}
}
}
[mcp_servers.kernel]
command = "uvx"
args = ["actenon-kernel"]
{
"mcpServers": {
"kernel": {
"args": [
"actenon-kernel"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"kernel": {
"args": [
"actenon-kernel"
],
"command": "uvx"
}
}
}
Authorize consequential AI agent actions before execution
MUJ428 verifies intent and evidence before consequential AI-agent actions.
AI Agent Firewall. Enforce signed authority before consequential actions. Free API key required.
Independent effect verification and signed receipts for consequential AI agent actions.
Pre-execution governance for AI agents. Validates tool calls before they execute.
Pre-action gate: verify before pay/trade/write/deploy. execute=true only on ALLOW.
Exact-action approval for consequential agent actions: request, track, and verify signed receipts.
Private, voluntary AI-assisted conciliation with human approval before consequential actions.
Answers built from our own checks of this server.