agent-lint

com.runningunattendedv0.1.0Updated Oct 4, 2026

Checks CLAUDE.md, AGENTS.md or a system prompt for 12 safeguards unattended agents need.

VerifiedStreamable HTTPWeb executableDeveloper ToolsAI & ML

Overview

AI-generated overview

Checks an agent instruction file such as CLAUDE.md, AGENTS.md, or a system prompt against 12 safeguards for unattended agents.

What it does
This remote MCP server reviews agent instruction files and system prompts for the safeguards that unattended agents need. According to its description, it evaluates the text against 12 such safeguards and reports what is present or missing. No individual tools are listed in the manifest, so the exact interface is not documented here.
When to use it
Worth adding when you write or maintain CLAUDE.md, AGENTS.md, or system prompts for agents that run without a human watching, and you want a quick check of whether the instructions cover the expected safeguards before relying on them.
Requirements
A remote streamable HTTP endpoint at runningunattended.com; no packages, environment variables, headers, or authentication are declared. The assistant needs network access to that endpoint.
Before you install
The instruction file or system prompt you submit is sent to a third-party remote service, so avoid including secrets or sensitive internal content. No authentication is declared, so consider what that means for who can reach the endpoint.

Installation

In SourceWeft

  1. Open agent-lint in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "agent-lint": {
      "type": "http",
      "url": "https://runningunattended.com/api/mcp"
    }
  }
}

Tools

0
Tool metadata has not been indexed yet.

Version history

1
  1. v0.1.0LatestOct 4, 2026