MCP.so
登录

Trustloop

@SMJAI

代码仓库

关于 Trustloop

Governance and audit layer for MCP tool calls. TrustLoop intercepts every AI agent action, logs it to a tamper-evident audit trail, blocks unsafe tools via kill-switch, and anchors proof to the blockchain. Drop it into any MCP setup in minutes, no code changes required.

#governance#audit#compliance#security#ai-agents#kill-switch#blockchain#mcp-proxy

基本信息

分类

开发工具

传输方式

stdio

发布者

SMJAI

提交者

Soji Joseph

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "trustloop": {
      "url": "https://trustloop-production.up.railway.app/sse?api_key=<YOUR_TRUSTLOOP_API_KEY>"
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Trustloop?

Trustloop is an MCP governance proxy that intercepts tool calls from AI agents, checking them against policy rules before execution. It logs every call and either allows or blocks it, providing oversight and compliance for teams using AI agents.

How to use Trustloop?

Add the server to your claude_desktop_config.json with the provided URL and your API key, then configure policy rules to govern agent tool usage. Get your API key at trustloop.live.

Key features of Trustloop

  • Kill-switch to block specific tools by name
  • Audit log capturing timestamp, tool, arguments, and status
  • Blockchain anchoring of audit log hashes for tamper-evident compliance
  • Multi‑tenant isolation of logs and governance rules
  • Built for EU AI Act and similar global mandates

Use cases of Trustloop

  • Prevent AI agents from executing destructive tool calls
  • Demonstrate regulatory compliance with auditable agent actions
  • Enforce governance policies across multiple agent customers
  • Maintain tamper‑evident records for external audits

FAQ from Trustloop

How does the kill-switch work?

You can block specific tools by name before they can cause harm. For example, delete_all_files is blocked by default.

What is the audit log?

Every tool call is captured with its timestamp, tool name, arguments, and status (allowed or blocked).

How does blockchain anchoring work?

A SHA-256 hash of the audit logs is recorded on‑chain, making logs tamper‑evident for compliance.

Is Trustloop multi‑tenant?

Yes, each customer gets isolated logs and governance rules.

What do I need to run Trustloop?

You need an API key from trustloop.live and to configure an MCP client (e.g., Claude Desktop) with the server URL.

评论

开发工具 分类下的更多 MCP 服务器

MCP Framework

QuantGeekDev

The Typescript MCP Framework

924

MCP Containers

metorial

Connect any AI model to 1200+ integrations (MCP, CLI, API)

3.3K

MCP-Bridge

SecretiveShell

A middleware to provide an openAI compatible endpoint that can call MCP tools

928

Sentry

modelcontextprotocol

Model Context Protocol Servers

88.1K

Minecraft MCP Server

yuniko-software

A Minecraft MCP Server powered by Mineflayer API. It allows to control a Minecraft character in real-time, allowing AI assistants to build structures, explore the world, and interact with the game environment through natural language instruction

642