mcpdirectory.dev
9,129 servers · updated 23 Sept

Agent Trust Receipt MCP server

community Active last commit 24 d 
{
  "mcpServers": {
    "ai-akari-agent-trust-receipt": {
      "type": "http",
      "url": "https://ai-akari.ai/mcp-trust"
    }
  }
}
file .mcp.json (in your project root)version 1.1.0

What it does

Agent Trust Receipt is a read-only remote MCP server that audits claimed outcomes from AI agents, classifying them as result, progress, insufficient_evidence, or rejected. It is intended for AI agents, agent marketplaces, and autonomous teams that need to verify a claimed outcome before accepting or acting on it.

  • Verify whether an AI agent's reported outcome constitutes a genuine result or lacks sufficient evidence before accepting it.
  • Retrieve the classification of a claim along with missing proof requirements and a recommended next verification action.
  • Integrate trust receipt auditing into an autonomous agent pipeline without authentication, using the free remote MCP endpoint.

Tools this server exposes

ToolDoes
audit_result_claimClassifies a submitted claim as result, progress, insufficient_evidence, or rejected, and returns missing proof and the next verification action.

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.