# roles

A execute tool on the iTerm MCP MCP server.

Agent View of the PolicyLayer registry record for `roles`. HTML page: https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/roles

## Facts

- Tool: `roles`
- Server: iTerm MCP (`research-developer/iterm-mcp-claude-agency`) — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency.md
- Homepage: https://github.com/research-developer/iterm-mcp-claude-agency
- Risk category: Execute (High risk)
- Registry record: grade F, 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": "roles",
    "arguments": {}
  }
}
```

## Why roles is rated High

Without a description, confidence is reduced. However, 'roles' on a terminal control server most likely manages role-based access control, which determines who can execute commands via the iTerm sessions. This makes it Execute-adjacent (controls what can be executed) rather than purely Read, though it could also be Write if it modifies role assignments.

From the tool's own definition: "Tool named 'roles' with empty description on an iTerm MCP server that controls terminal sessions and command execution."

## Use case

AI agents invoke roles to trigger actions in iTerm MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.

## Recommended policy (PolicyLayer)

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

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "roles": {
      "limits": [
        {
          "counter": "roles_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on iTerm MCP (15)

- `delegate` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/delegate.md
- `managers` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/managers.md
- `orchestrate` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/orchestrate.md
- `services` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/services.md
- `sessions` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/sessions.md
- `telemetry` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/telemetry.md
- `wait_for` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/wait-for.md
- `workflows` — Execute — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/workflows.md
- `agents` — Other — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/agents.md
- `bus` — Other — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/bus.md
- `feedback` — Other — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/feedback.md
- `memory` — Other — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/memory.md
- `messages` — Other — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/messages.md
- `subscribe` — Other — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/subscribe.md
- `teams` — Other — https://policylayer.com/tools/research-developer-iterm-mcp-claude-agency/teams.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=research-developer-iterm-mcp-claude-agency · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/research-developer-iterm-mcp-claude-agency