Privacy settings are useful when they answer a real question at the moment a task needs an answer. A single “private mode” toggle is not enough for a desktop agent that can read code, call tools, and connect to model providers. Nerva should expose a small set of understandable controls.

Settings with practical meaning

Setting What it should control
Local-only Prevent provider API requests for the task
Ask before cloud Show context and destination before sending
Workspace paths Limit file reads and writes to selected folders
Secret handling Block or redact likely credentials
Tool approvals Ask before writes, deletes, commands, and network
History retention Control how long task records remain locally

These settings should be visible in the task, not buried in a settings page the user never sees. A saved policy can reduce repeated prompts, but a scope change should still trigger a new check.

Explain exceptions

If a local model cannot handle a task and Nerva suggests a cloud route, the explanation should include what would leave the device and why. If a requested operation is denied, show which rule blocked it and how to narrow the task. Good privacy UX makes the safe choice understandable rather than simply refusing.

For developers working with private repositories or fintech code, those details matter more than a generic privacy slogan. Trust grows when the product shows the boundary every time it crosses one.

The useful design question

The useful question here is not whether an AI agent can produce an impressive demo. It is whether the workflow stays understandable when the task is incomplete, the repository is messy, or the model is wrong. For this part of Nerva, the focus is giving users privacy controls they can actually understand. That means each state should have a visible owner, a small input, and a result that another part of the system can inspect.

A concrete scenario

Consider choosing local-only context for credentials and a reviewed cloud route for sanitized code. A good Nerva run would first state the scope, identify the files or operation it needs, and show the next proposed step. It would keep the original request separate from model-generated instructions. If an action needs approval, the user would see what will happen before it happens. Afterward, the task history would retain the command, result, and verification status instead of reducing everything to a confident paragraph.

This approach also makes a failure useful. A denied permission, unavailable provider, missing package, or changed file is information about the task state. Nerva should preserve it and explain the smallest recovery option. It should not silently broaden the request, retry a side effect, or claim completion because a model supplied plausible-looking text.

What I would test first

For an early implementation, I would test three paths: a normal successful run, a denied action, and an interrupted run resumed after the workspace changes. The key test is whether every setting explains its practical consequence. In each path, the interface should make clear what Nerva knows, what it proposed, what actually ran, and what remains unverified.

Useful acceptance checks include:

  • the user can inspect the selected context before model execution;
  • sensitive paths and credentials are excluded or redacted;
  • every side effect has a narrow permission decision;
  • failures remain attached to the step that caused them;
  • a final answer links to evidence instead of implying that unrun work succeeded.

Nerva is still an upcoming project, so these are design and implementation goals rather than claims that a finished product already provides them. That distinction matters: a trustworthy agent should earn confidence through observable behavior, not through a feature list.

A small review checklist

A practical review should end with a small decision, not a vague promise. Write down the scope, the expected result, the permission required, and the check that will confirm it. Then try the same flow with a refusal or missing dependency. If the interface explains that case clearly, the design is ready for a more realistic prototype; if it hides the problem, adding another feature will only make the workflow harder to trust.

The design is informed by the boundaries I have had to think through in CryptoEx and SentinelFi: backend work becomes easier to review when context, permissions, side effects, and verification are explicit.