Infraveil Agent Guard for VS Code

io.github.infraveilhq/agent-guard

Gate an AI agent's destructive actions behind human approval, with a tamper-evident log.

client:VS Code transport:stdio runtime:pypi

Install Infraveil Agent Guard in VS Code

.vscode/mcp.json

{
  "inputs": [
    {
      "type": "promptString",
      "id": "infraveil-guard-threshold",
      "description": "INFRAVEIL_GUARD_THRESHOLD",
      "password": true
    },
    {
      "type": "promptString",
      "id": "infraveil-guard-mode",
      "description": "INFRAVEIL_GUARD_MODE",
      "password": true
    },
    {
      "type": "promptString",
      "id": "infraveil-guard-home",
      "description": "INFRAVEIL_GUARD_HOME",
      "password": true
    }
  ],
  "servers": {
    "agent-guard": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "infraveil-guard"
      ],
      "env": {
        "INFRAVEIL_GUARD_THRESHOLD": "${input:infraveil-guard-threshold}",
        "INFRAVEIL_GUARD_MODE": "${input:infraveil-guard-mode}",
        "INFRAVEIL_GUARD_HOME": "${input:infraveil-guard-home}"
      }
    }
  }
}

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

Infraveil Agent Guard in other clients

Infraveil Agent Guard MCP server