mcpdirectory.dev
9,129 servers · updated 23 Sept

Goodeye MCP server

community Active last commit 8 d
{
  "mcpServers": {
    "goodeye": {
      "type": "http",
      "url": "https://mcp.goodeye.dev/mcp"
    }
  }
}
file .mcp.json (in your project root)version 0.19.2

What it does

Goodeye is a system for managing AI workflow skills (markdown runbooks) and verifiers (output checks) that can be hosted, synced across machines, and shared with team members. It's for developers who want to maintain consistent AI agent behavior across multiple environments and collaborators.

  • Publish a skill to a private hosted account, then sync it automatically across your laptop, CI, and team members' machines so everyone runs the same current version.
  • Create a verifier that checks AI output (format validation, schema compliance, or semantic quality via LLM judge) and attach it to a skill so every run validates against that exact version.
  • Share a skill with a teammate by granting them access, and the verifiers it references come along automatically at the same pinned version.

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.