MCP Server MCP

TAR Engine

By github 👁 1 views ▲ 0 votes

Audit AI skill safety before you ship. Static, semantic, adversarial, supply chain scans.

Homepage Verified source GitHub

Install

uvx tar-engine

Configuration Example

{
  "remotes": [],
  "packages": [
    {
      "registryType": "pypi",
      "identifier": "tar-engine",
      "version": "0.3.3",
      "runtimeHint": "uvx",
      "transport": {
        "type": "stdio"
      },
      "packageArguments": [
        {
          "description": "Install the tar-engine wheel that provides the tar-engine-mcp executable.",
          "value": "tar-engine",
          "type": "positional",
          "name": "--from"
        },
        {
          "description": "Run the MCP server entry point exposed by the tar-engine wheel.",
          "value": "tar-engine-mcp",
          "type": "positional",
          "name": "tar-engine-mcp"
        }
      ],
      "environmentVariables": [
        {
          "description": "Override the audit backend. Default is the hosted playground at https://tarai.dev. Set to http://localhost:8765 (or your own tar-engine deployment) to self-host and keep SKILL.md content on your machine.",
          "name": "TAR_ENGINE_URL"
        },
        {
          "description": "OpenAI API key that unlocks the semantic LLM and adversarial prompt-fuzz layers. Static and supply-chain layers run without it.",
          "isSecret": true,
          "name": "TAR_ENGINE_BYOK_OPENAI_KEY"
        },
        {
          "description": "Anthropic API key alternative to the OpenAI key for the semantic and adversarial layers.",
          "isSecret": true,
          "name": "TAR_ENGINE_BYOK_ANTHROPIC_KEY"
        }
      ]
    }
  ]
}

mcp model-context-protocol pypi