verdict_command_gate

Free sample. Before a coding agent runs a shell command, ask whether to allow, review or block it: root deletion, download and run, privilege escalation, secret exfiltration. Pass the human intent when you know it. Not a sandbox: it reads the command, it does not run it.

Argument	Type	Description
intent	string	Optional. What the human asked the agent to do.
command*	string	The shell command the agent wants to run.

Input schema (JSON)

{
  "type": "object",
  "required": [
    "command"
  ],
  "properties": {
    "intent": {
      "type": "string",
      "description": "Optional. What the human asked the agent to do."
    },
    "command": {
      "type": "string",
      "description": "The shell command the agent wants to run."
    }
  }
}

verdict_firewall

Free sample. Before an agent acts on untrusted content (pay, send, delete, click, reply), give the human goal, the content it read and the action it proposes. Returns allow, review or block: it catches actions that follow orders hidden in the content.

Argument	Type	Description
goal*	string	The objective the human gave.
action*	string	The action the agent wants to run (tool call, command, payment).
content*	string	The untrusted content the agent read.

Input schema (JSON)

{
  "type": "object",
  "required": [
    "goal",
    "content",
    "action"
  ],
  "properties": {
    "goal": {
      "type": "string",
      "description": "The objective the human gave."
    },
    "action": {
      "type": "string",
      "description": "The action the agent wants to run (tool call, command, payment)."
    },
    "content": {
      "type": "string",
      "description": "The untrusted content the agent read."
    }
  }
}

verdict_guard

Free sample. Before your agent reads untrusted text (web page, email, document, tool output), check it for prompt injection, data exfiltration requests, secrets and toxic content. Returns allow, review or block with scores and reasons. Use it on text you did not write. Do not use it to judge an action: use verdict_firewall.

Argument	Type	Description
text*	string	The untrusted text to examine (4000 characters maximum in free samples).

Input schema (JSON)

{
  "type": "object",
  "required": [
    "text"
  ],
  "properties": {
    "text": {
      "type": "string",
      "description": "The untrusted text to examine (4000 characters maximum in free samples)."
    }
  }
}

verdict_mcp_vet

Free sample. Before installing an MCP server, skill or tool: is its description poisoned (hidden orders to the agent, excessive permissions, look-alike names)? Returns allow, review or block with reasons.

Argument	Type	Description
name*	string	Name of the tool or server.
tools	array	Optional list of { name, description } of the functions it exposes.
description*	string	The description text it exposes to the agent.
permissions	array	Optional permissions requested, for example filesystem:write, network, shell.

Input schema (JSON)

{
  "type": "object",
  "required": [
    "name",
    "description"
  ],
  "properties": {
    "name": {
      "type": "string",
      "description": "Name of the tool or server."
    },
    "tools": {
      "type": "array",
      "items": {
        "type": "object"
      },
      "description": "Optional list of { name, description } of the functions it exposes."
    },
    "description": {
      "type": "string",
      "description": "The description text it exposes to the agent."
    },
    "permissions": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Optional permissions requested, for example filesystem:write, network, shell."
    }
  }
}

Verdict L402

Try it

verdict_command_gate

Tools (4)

Verdict L402

Try it

verdict_command_gate

Tools (4)