# create_agent

A write tool on the AgentPay MCP server.

Agent View of the PolicyLayer registry record for `create_agent`. HTML page: https://policylayer.com/tools/advaitgore-agent-payment/create-agent

## Facts

- Tool: `create_agent`
- Server: AgentPay (`advaitgore/agent_payment`) — https://policylayer.com/tools/advaitgore-agent-payment.md
- Homepage: https://github.com/advaitgore/agent_payment
- Risk category: Write (Medium risk)
- Registry record: grade D, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "create_agent",
    "arguments": {}
  }
}
```

## Why create_agent is rated Medium

This tool creates a new agent account/identity on a financial authorization system. While not immediately destructive or moving funds itself, creating agents on a spending authorization platform is a write operation with high severity due to the financial exposure it enables—a newly created agent could be used to establish unauthorized spending mandates or circumvent controls.

From the tool's own definition: "Tool name 'create_agent' and server context (AgentPay authorization layer) indicate creation of a new agent entity. Server description emphasizes financial spending controls and real-money transaction authorization."

## Use case

AI agents use create_agent to create or update resources in AgentPay, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your AgentPay environment.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches AgentPay:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "create_agent": {
      "limits": [
        {
          "counter": "create_agent_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on AgentPay (8)

- `authorize_purchase` — Financial — https://policylayer.com/tools/advaitgore-agent-payment/authorize-purchase.md
- `create_account` — Financial — https://policylayer.com/tools/advaitgore-agent-payment/create-account.md
- `create_mandate` — Financial — https://policylayer.com/tools/advaitgore-agent-payment/create-mandate.md
- `update_mandate` — Financial — https://policylayer.com/tools/advaitgore-agent-payment/update-mandate.md
- `get_audit_log` — Read — https://policylayer.com/tools/advaitgore-agent-payment/get-audit-log.md
- `get_mandate` — Read — https://policylayer.com/tools/advaitgore-agent-payment/get-mandate.md
- `get_spending_summary` — Read — https://policylayer.com/tools/advaitgore-agent-payment/get-spending-summary.md
- `rotate_agent_key` — Write — https://policylayer.com/tools/advaitgore-agent-payment/rotate-agent-key.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=advaitgore-agent-payment · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/advaitgore-agent-payment