psh.now is answering right now. Last checked 8 min ago. It exposes 1 tools. Last commit 12 Nov 2026.
Human approval for AI agents: one ask, one person, one boolean.
We read the source, 15 h ago · tools taken from the live server · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
/(rm\s+-[a-z]*r[a-z]*f[a-z]*|--force\b|\bsudo\b|git\s+push\b[^\n]*(?:--force|\s-f\b)|DROP\s+TABLE|TRUNCATE\s+TABLE|DELETE\s+FROM(?![\s\S]*WHERE)|\bmkfs\b|\bdd\b[^\n]*of=\/dev\/|chmod\s+777|kubectl\s+delete|terraform\s+destroy|\|\s*(?:ba|z|d)?sh\b|shutdown\b|reboot\b)/gi;
el.innerHTML = svg(el.dataset.icon, Number(el.dataset.iconSize || 16));
<script defer src="https://ps.abslm.nl/script.js" data-website-id="9d2c0261-73aa-4550-84a2-ee572a4c5934"></script>
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 8 min ago.
claude mcp add psh-now --transport http https://psh.now/mcp
https://psh.now/mcp
[mcp_servers.psh-now]
url = "https://psh.now/mcp"
{
"mcpServers": {
"psh-now": {
"url": "https://psh.now/mcp"
}
}
}
{
"servers": {
"psh-now": {
"type": "http",
"url": "https://psh.now/mcp"
}
}
}
Read directly from the server with tools/list, grouped by what they act on.
If a tool disappears, we record the date.
psh_ask
| URL | Transport | State | Latency | Checked |
|---|---|---|---|---|
| https://psh.now/mcp | streamable-http | answering | 287 ms | 8 min ago |
Human approval gate for AI agents: risk-scored requests that only a person can approve
Human approval for AI agents: a named person approves or rejects, silence is not consent.
Prepaid virtual cards for AI agents: one-time cards, spend caps, human approvals.
Set up human approval for your AI agents' tool calls.
Ask a human before your agent acts: approval requests decided on a phone.
Preventive human-approval write-gate for AI agents: writes commit only after a human approves.
Give any AI agent a way to ask a person — for approval, a decision, an answer or a review.
Impri MCP server — human-in-the-loop approval inbox for AI agents
Answers built from our own checks of this server.