← All help topics

Privacy and security

SolutionMCP gives AI apps real access to your computer, so it's built around three ideas:

  1. Ask before risky actions. Deleting, overwriting and running commands need your OK, unless you decide otherwise.
  2. Limit what can be reached. File tools only work in the folders you allow, and keys, passwords and private data stay blocked.
  3. Keep a record. Every tool call is written to the activity log on your computer.

What leaves your computer

SolutionMCP itself sends nothing anywhere, except:

  • the requests your tools make, like reading a web page you asked about
  • the results it gives your AI app, which the app usually sends on to its AI service
  • the update check, which reads the list of releases from solution-mcp.com
  • licenses: using a license on this computer, checking for a renewal about once a week, asking how the license stands about once a day (for the Account page), and renaming or freeing a computer or changing your details there. These send the license key or the license's ID, this computer's ID, the version of SolutionMCP and what you changed, and nothing about your tools, files or what your AI apps did. Using a license also sends this computer's name, so the license's computers can be told apart. The license itself is checked on your computer, offline. The Account page's activity numbers come from this computer's own log and aren't sent.

So don't have the AI read a file into a conversation you wouldn't paste into it yourself.

What protects you

  • Approvals for risky tools. See Approvals.
  • Folder limits and blocked locations for file tools. See Folders the AI can use.
  • SolutionMCP protects itself. The AI can't change SolutionMCP's settings, can't read your secrets, and can't sign in to the dashboard.
  • Safe values. What the AI sends can't sneak commands into tools: it's passed separately or quoted.
  • Blocked commands. Some terminal commands, like sudo and shutdown, are always refused.

What it can't protect against

  • Tricks hidden in content. A web page, email or document can contain instructions aimed at the AI. Keep approvals on for anything that deletes or overwrites.
  • Approving the wrong thing. Read approval questions: they show what will run.
  • AI agents with their own terminal. Coding agents that run commands act as you, outside SolutionMCP. Keep their own approvals on.
  • Trusted mode and broad "Always run" rules turn the questions off. Use them only where you're comfortable with that.