# disconnect

Disconnect from current Redis instance

Agent View of the PolicyLayer registry record for `disconnect`. HTML page: https://policylayer.com/tools/adamhancock-bullmq-mcp/disconnect

## Facts

- Tool: `disconnect`
- Server: BullMQ MCP Server (`adamhancock/bullmq-mcp`) — https://policylayer.com/tools/adamhancock-bullmq-mcp.md
- Homepage: https://github.com/adamhancock/bullmq-mcp
- Risk category: Write (Medium 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": "disconnect",
    "arguments": {}
  }
}
```

## Why disconnect is rated Medium

Disconnecting from a Redis instance is a state management operation that terminates a connection but does not create, delete, or permanently alter stored data. It is reversible—the user can reconnect. While it has operational consequences (queued operations may be affected), it does not irreversibly destroy data or execute arbitrary code. Write is the appropriate category for reversible state modifications.

From the tool's own definition: "Tool name is 'disconnect' and description states it 'Disconnect from current Redis instance'. This modifies the connection state (closes/terminates a connection), which is a reversible change to the application's runtime state rather than data persistence."

## Use case

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

## Recommended policy (PolicyLayer)

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

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

## Other tools on BullMQ MCP Server (19)

- `clean_queue` — Destructive — https://policylayer.com/tools/adamhancock-bullmq-mcp/clean-queue.md
- `drain_queue` — Destructive — https://policylayer.com/tools/adamhancock-bullmq-mcp/drain-queue.md
- `remove_job` — Destructive — https://policylayer.com/tools/adamhancock-bullmq-mcp/remove-job.md
- `pause_queue` — Execute — https://policylayer.com/tools/adamhancock-bullmq-mcp/pause-queue.md
- `resume_queue` — Execute — https://policylayer.com/tools/adamhancock-bullmq-mcp/resume-queue.md
- `retry_job` — Execute — https://policylayer.com/tools/adamhancock-bullmq-mcp/retry-job.md
- `switch_connection` — Other — https://policylayer.com/tools/adamhancock-bullmq-mcp/switch-connection.md
- `get_job` — Read — https://policylayer.com/tools/adamhancock-bullmq-mcp/get-job.md
- `get_job_logs` — Read — https://policylayer.com/tools/adamhancock-bullmq-mcp/get-job-logs.md
- `get_jobs` — Read — https://policylayer.com/tools/adamhancock-bullmq-mcp/get-jobs.md
- `list_connections` — Read — https://policylayer.com/tools/adamhancock-bullmq-mcp/list-connections.md
- `list_queues` — Read — https://policylayer.com/tools/adamhancock-bullmq-mcp/list-queues.md
- `query_dead_letter_queue` — Read — https://policylayer.com/tools/adamhancock-bullmq-mcp/query-dead-letter-queue.md
- `stats` — Read — https://policylayer.com/tools/adamhancock-bullmq-mcp/stats.md
- `add_job` — Write — https://policylayer.com/tools/adamhancock-bullmq-mcp/add-job.md
- `add_job_log` — Write — https://policylayer.com/tools/adamhancock-bullmq-mcp/add-job-log.md
- `connect` — Write — https://policylayer.com/tools/adamhancock-bullmq-mcp/connect.md
- `move_failed_jobs_to_dlq` — Write — https://policylayer.com/tools/adamhancock-bullmq-mcp/move-failed-jobs-to-dlq.md
- `promote_job` — Write — https://policylayer.com/tools/adamhancock-bullmq-mcp/promote-job.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=adamhancock-bullmq-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/adamhancock-bullmq-mcp