Disclaimer

Last updated 15 July 2026

Metroit X uses AI models to understand your requests and decide what actions to take. Please read this before you let the agent make changes on your machine.

AI responses, code, and commands may be inaccurate, incomplete, or unsafe. The agent can be confidently wrong. Always review what it proposes — especially anything destructive (deleting files, overwriting data, force-pushing, dropping a database, running a script you don’t fully understand) — before letting it run on a machine that matters to you.

1. No guarantee of accuracy

Responses are generated by third-party AI models (Google Gemini, Anthropic Claude, or models routed through OpenRouter, depending on what you’ve selected). These models can hallucinate facts, write code with bugs or security issues, and misunderstand your intent. We don’t guarantee that any Suggestion is correct, complete, safe to run, or fit for any particular purpose.

2. The agent acts on your real machine

Unlike a chatbot that only produces text, Metroit X’s agent can actually read, write, and delete files, and execute shell commands on the machine where you’ve installed the Plugin. Our permission engine reviews commands and flags or blocks many risky patterns before they reach your machine — but this is a best-effort safety layer, not a guarantee. New or unusual commands can slip past any automated check. You are the last line of defense for anything that runs on your own computer.

3. Keep your own backups

Before using the agent on anything you can’t afford to lose — important code, production data, personal files — make sure you have your own backups or version control in place. We are not responsible for data loss, corruption, or damage resulting from actions the agent takes at your direction.

4. No liability for AI or agent actions

To the maximum extent permitted by law, we are not liable for any loss or damage — including lost data, lost revenue, or system downtime — arising from Suggestions generated by the Service or from commands/file changes the agent carries out on your machine, whether or not those actions were flagged by our safety checks. See the Limitation of Liability section in our Terms of Service for the full terms.

5. Not professional advice

Nothing the agent produces (code, security recommendations, configuration changes, or anything else) constitutes professional, legal, financial, or security advice. Use your own judgment, especially for production systems and anything security-sensitive.

6. Service availability

The Service depends on third-party AI providers and infrastructure we don’t control, and may occasionally be slow, degraded, or unavailable.
We don’t guarantee uninterrupted or error-free operation, though we make reasonable efforts to keep things running well.

7. Questions

If anything here is unclear, email info@dremora.co before you connect the Plugin to a machine you’re worried about.