Raposa Aval — human approval runs on your own machine — the client starts it, so there is no endpoint to ping. 632 installs a week from pypi. Last commit 3 Sep 2026.
Human approval for AI agents: a named person approves or rejects, silence is not consent.
Today is the operative word: we check Raposa Aval — human approval 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 raposa-mcp -- uvx raposa-mcp
{
"mcpServers": {
"raposa-mcp": {
"args": [
"raposa-mcp"
],
"command": "uvx"
}
}
}
[mcp_servers.raposa-mcp]
command = "uvx"
args = ["raposa-mcp"]
{
"mcpServers": {
"raposa-mcp": {
"args": [
"raposa-mcp"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"raposa-mcp": {
"args": [
"raposa-mcp"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
RAPOSA_API_KEY (Raposa Aval API key (free sandbox key at https://raposa.group/start/)).
The author declared them in the registry entry; get the values from the project itself.
Human approval gate for AI agents. Critical actions approved or rejected by a human via Telegram.
Virtual Visa cards for AI agents. JIT card issuance, human-approved, MCP-native.
Preventive human-approval write-gate for AI agents: writes commit only after a human approves.
Governed, self-hosted memory for AI agents: writes queue until a named person approves.
A real email inbox for AI agents: human-approved sends, untrusted-by-default inbound.
Intent-bound action authorization for AI agents: policy, human approval, and a signed audit trail.
Your real mailbox and calendar for your agent (M365/IMAP); a human approves every send.
Agent Relay MCP: invite-based channels, agent contracts and per-side human approvals.
Answers built from our own checks of this server.