SNcodebeta
DocsPricingDownload

The Review Gate

SNcode doesn't run AI-generated code behind your back. Every script stops at the same review gate: you see a plain-language description and the exact code, and nothing touches your instance until you approve it.

Review-first execution

The AI proposes; you approve. When SNcode generates a script, it presents three things you control, in order:

  1. 1

    Describe

    A natural-language summary of what the script will do and which tables or records it touches.
  2. 2

    Inspect

    The full generated code, exactly as it will run. Nothing hidden, nothing minified.
  3. 3

    Approve and run

    When you approve, the script runs on your connected instance exactly as shown. Nothing executes until you say so.

A review gate, not a sandbox

SNcode does not run scripts in an isolated environment. An approved script executes directly on the connected instance with that account's full permissions, so connecting an admin account gives an approved script admin-level reach. Connect a least-privilege account, and try changes on a PDI or sub-production instance before production.

Turning the gate off

The gate is on for every chat by default. A chat's settings include a Skip confirmation toggle; while it is on, scripts and actions in that chat run as soon as the AI proposes them, without waiting for your approval. Use it only on instances where you'd accept that, and switch it off again when you're done.

Why it matters

  • Large language models can produce incorrect or incomplete code, so a human review gate is on by default
  • You keep an auditable record of what was proposed versus what was run
  • Admins can connect a client instance knowing no code executes without explicit approval unless they turn the gate off

AI output is not automatically trusted

All AI-generated code, scripts, and recommendations must be reviewed by a qualified human before execution or deployment. The review gate exists to make that review the default step.

Execution history

Script executions and their outputs are recorded locally on your device, so you can trace exactly what ran against which instance and when. That history is useful for audits and for resuming work across sessions.

Next

The AI reasons over more than tables. Next, see how attaching CSV, PDF, and Excel files lets it transform your own data, behind the same review gate.