Quirna runs on your own machine — the client starts it, so there is no endpoint to ping. 37 installs a week from npm. Last commit 27 Sep 2026.
Ask a human before your agent acts: approval requests decided on a phone.
Today is the operative word: we check Quirna 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 -- npx -y @quirna/mcp
{
"mcpServers": {
"mcp": {
"args": [
"-y",
"@quirna/mcp"
],
"command": "npx"
}
}
}
[mcp_servers.mcp]
command = "npx"
args = ["-y", "@quirna/mcp"]
{
"mcpServers": {
"mcp": {
"args": [
"-y",
"@quirna/mcp"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"mcp": {
"args": [
"-y",
"@quirna/mcp"
],
"command": "npx"
}
}
}
This one needs environment variables set before it will start:
QUIRNA_API_KEY (Key of a System registered in the Quirna Console (Users → Systems).), QUIRNA_REQUESTER_ID (Who the Console shows as having asked. Defaults to "mcp".), QUIRNA_REQUESTER_NAME (Display name for the requester. Defaults to "AI agent".), QUIRNA_ENVIRONMENT (Environment used when a tool call names none. Empty means the org default.), QUIRNA_WAIT_MS (How long a tool call waits for a decision before returning pending. Defaults to 90000.), QUIRNA_BASE_URL (Override the Quirna API URL. Leave unset.).
The author declared them in the registry entry; get the values from the project itself.
logi Agent Approval Gate — human approval on the phone before high-risk agent actions.
Puts Approve/Deny on your phone before an agent acts, and waits for your fingerprint.
The body an agent asks before it acts: decide, approve, and keep a signed record on your machine.
Your agent posts, schedules, queues and threads posts on X, with a human approval queue.
Human-in-the-loop approval gateway for AI agents over MCP: request, decide, approve.
Deterministic allow/require_approval/deny verdicts for agent actions, before they happen.
LoopPause MCP server — pause agent execution and route human approval requests
Your agent asks you a question on your phone, waits for the answer, then resumes.
Answers built from our own checks of this server.