Verified

Trmm MCP Server

TacticalRMM managed machines data access and fleet command execution control.

Data last scanned today · Reviewed today

45 8 dangerous tools28 tools★ 2PythonAGPL-3.0

Overview

This server integrates TacticalRMM with AI assistants, providing visibility into managed fleet data including agents, service statuses, event logs, and software inventory. It processes 28 tools in total, allowing for deep diagnostic queries and reporting across your infrastructure. The server includes execution capabilities for rebooting machines, managing services, and running scripts. These actions are protected by an approval gate requiring manual confirmation via browser or shell before any command runs on managed endpoints.

Our verdict

This server carries a high-risk rating with a security score of 30/100. The elevated risk stems from the ability to execute remote commands, combined with the fact that the project is community-maintained rather than official vendor software. Users should exercise extreme caution, ensuring the audit logs are monitored closely and that the read-only or execution permissions are strictly scoped to the underlying TacticalRMM role.

Supports zero direct dependencies for reduced supply-chain exposure
Provides mandatory JSONL audit logging for all tool calls
Implements triple-layer protection for read-only operation
Includes automated truncation of large responses to preserve context

Carries a high-risk classification due to command execution capabilities
Maintained by an individual contributor with a low bus factor
Lacks an official vendor-supported status
Authentication and read-only implementation details remain unverified

Tools

Tool	Description	Risk
trmm_fleet_overview	Returns counts plus the offline, needs-reboot, failing-check, and pending-patch lists.	low
trmm_list_agents	Lists agents with status, filterable by client, site, type, online state, or hostname.	low
trmm_get_agent	Returns full detail for one machine with heavy sections omitted by default.	low
trmm_agent_services	Lists Windows services, filterable by name and state.	low
trmm_agent_processes	Lists live processes sorted by memory or CPU.	low
trmm_agent_event_log	Returns the Windows event log, filterable by level and text.	low
trmm_agent_checks	Returns configured checks and their status.	low
trmm_agent_history	Returns past command and script runs with their output.	low
trmm_agent_software	Returns installed software inventory.	low
trmm_agent_windows_updates	Returns Windows patch status.	low

Compatibility

Client	Local	Docker	Remote	Read-only
ChatGPT
Claude Desktop
Cursor
VS Code
Windsurf

Frequently asked questions

›Which TacticalRMM versions are supported?

This server is built and verified against TacticalRMM version 1.5.1.

›How does the 'elevate' mode work for executing commands?

In elevate mode, the assistant is blocked from executing any command until you approve the specific action in your browser via password and TOTP, or from a server shell.

›How does the server protect against large API responses?

Since TacticalRMM does not support pagination, the server projects, caps, and truncates responses to prevent them from overwhelming the model's context window.

›What security measures are in place to ensure read-only mode?

Read-only access is enforced across three layers: execution tools are not registered, the HTTP client rejects non-read requests, and the associated API key is restricted to a role with no execution permissions.

›Are there built-in safeguards against destructive commands in 'command' mode?

Yes, destructive patterns like 'rm -rf /' or 'format c:' are refused client-side, and you can optionally define an agent allowlist to restrict which machines can be accessed.

›Can I rotate the API keys used by the server?

Yes, you can rotate keys by running the provisioning script provided at /opt/trmm-mcp/provision_trmm_accounts.py with the --rotate flag.