MCP.so
Sign In
A

Agent Receipts

@webaesbyamin

Repository

About Agent Receipts

Cryptographic accountability for AI agents. Ed25519-signed receipts for every MCP tool call — constraints, chains, AI judgment, invoicing, local dashboard.

#ed25519#receipts#audit#accountability#agents#typescript#mcp#signing

Basic information

Category

AI & Agents

Transports

stdio

Publisher

webaesbyamin

Submitted by

Amin Suleiman

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "agent-receipts": {
      "command": "npx",
      "args": [
        "@agent-receipts/mcp-server"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Agent Receipts?

Agent Receipts generates a signed, verifiable receipt for every action an AI agent takes. Every receipt is Ed25519 signed using a locally-generated key pair — no server, no accounts, no data leaving your machine. It is designed for developers using MCP-compatible clients like Claude Desktop, Cursor, or VS Code.

How to use Agent Receipts?

Agent Receipts works with any MCP-compatible client (Claude Desktop, Cursor, VS Code). It exposes 14 MCP tools including track_action, verify_receipt, judge_receipt, and generate_invoice. A live demo is available at https://agent-receipts-web.vercel.app.

Key features of Agent Receipts

  • Ed25519 signed receipts with local key generation
  • 14 MCP tools for tracking, verifying, and judging
  • Constraint evaluation (max latency, max cost, min confidence)
  • Receipt chains for multi-step agent workflows
  • AI judgment system with rubric-based quality scoring
  • Invoice generation from receipt data
  • Local SQLite storage and Next.js dashboard

Use cases of Agent Receipts

  • Tracking every action an AI agent takes with a signed receipt
  • Verifying receipt authenticity and constraint satisfaction
  • Judging agent output quality using rubric-based scoring
  • Generating invoices from receipt data for billing or auditing
  • Auditing multi-step agent workflows via receipt chains

FAQ from Agent Receipts

Does Agent Receipts require a server or account?

No. Everything runs locally — no server, no accounts, and no data leaves your machine.

What signing algorithm is used?

Receipts are Ed25519 signed using a locally-generated key pair.

What data storage does Agent Receipts use?

It stores receipts and related data in a local SQLite database.

Which clients are supported?

It works with Claude Desktop, Cursor, VS Code, and any MCP-compatible client.

How many tools does Agent Receipts provide?

It provides 14 MCP tools, including track_action, verify_receipt, judge_receipt, generate_invoice, and more.

Comments

More AI & Agents MCP servers

1MCP - One MCP Server for All

1mcp-app

A unified Model Context Protocol server implementation that aggregates multiple MCP servers into one.

468

meGPT - upload an author's content into an LLM

adrianco

Code to process many kinds of content by an author into an MCP server

287

1Panel

1Panel-dev

🔥 1Panel is a modern, open-source VPS control panel — and the only one with native AI agent support. Run Ollama models, deploy OpenClaw agents, and manage your entire server stack from one clean web interface.

36.1K

MCP-LLM Bridge

patruff

Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools

976

Sequential Thinking Multi-Agent System (MAS)

FradSer

An advanced sequential thinking process using a Multi-Agent System (MAS) built with the Agno framework and served via MCP.

306