The first version of an AI agent does not need to control every application. It needs one trustworthy workflow that works repeatedly. For Nerva, that means starting with local repository tasks, clear permissions, focused context, and verification before expanding into more ambitious automation.

A sensible first milestone

The first vertical slice could support:

  1. Select a workspace.
  2. Ask a local model to explain or plan a task.
  3. Show the selected context.
  4. Propose a diff or a read-only command.
  5. Ask before writing or executing.
  6. Run one focused verification step.
  7. Store the task result locally.

That slice proves the central promise without needing a huge plugin ecosystem. It also creates a foundation for the Rust permission engine, SQLite task history, model adapters, and future browser or document tools.

What can wait

Advanced autonomous loops, dozens of integrations, automatic cloud fallback, and unrestricted browser control should come later. Each adds a new failure and privacy surface. A smaller system with excellent state visibility is more useful than a broad demo that cannot explain what it did.

The roadmap should be measured by trustworthy completed workflows, not by the number of tools in a menu. If Nerva can help inspect a real backend repository, prepare a reviewable change, and prove the focused test result, it has already earned a meaningful place in a developer’s toolkit.

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 shipping a small trustworthy vertical slice first. 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 combining local context, one safe tool, approval, and verification before adding many integrations. 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 which promise can be demonstrated end to end rather than listed on a roadmap. 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.