mcpdirectory.dev
9,129 servers · updated 23 Sept

Raposa MCP server

community Active last commit 20 d
{
  "mcpServers": {
    "raposa": {
      "command": "uvx",
      "args": [
        "raposa-mcp",
        "raposa-mcp"
      ],
      "env": {
        "RAPOSA_API_KEY": "<raposa_api_key>"
      }
    }
  }
}
file .mcp.json (in your project root)install uvx raposa-mcpversion 0.1.1

What it does

Raposa is an MCP server that lets AI agents request human approval for actions, with explicit approve/reject decisions recorded in an audit log. It's for teams that need named person approval before agents execute sensitive operations, with EU hosting and compliance documentation available.

  • Request explicit human approval before an agent performs a high-risk action like modifying production data or sending communications.
  • Create async approval requests that notify team members by email with signed approve/reject links, receiving the decision later via webhook or polling.
  • Maintain an auditable record of who approved or rejected each agent action, sealed in a hash-chained log for regulatory compliance.

Tools this server exposes

ToolDoes
request_human_approvalCreates an approval request and waits for a human decision; returns approved status and the name of who decided, with timeout handling.
create_approvalCreates an approval request and returns immediately with an ID; the decision arrives asynchronously via webhook or polling.
get_approvalRetrieves the decision status for a specific approval request by ID.

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.