Skip to content
mcp/harbor
VS Code

agentshield mcp for VS Code

io.github.dl-eigenart/agentshield-mcp

Detect prompt injection, jailbreak, and social-engineering attacks in LLM agents.

client:VS Code transport:stdio runtime:npm

Install agentshield mcp in VS Code

.vscode/mcp.json 

{
  "inputs": [
    {
      "type": "promptString",
      "id": "agentshield-api-key",
      "description": "AGENTSHIELD_API_KEY",
      "password": true
    }
  ],
  "servers": {
    "agentshield-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@eigenart/agentshield-mcp"
      ],
      "env": {
        "AGENTSHIELD_API_KEY": "${input:agentshield-api-key}"
      }
    }
  }
}

VS Code prompts for each input the first time the server starts, so no secret is written to the file. VS Code docs

agentshield mcp in other clients

agentshield mcp MCP server