“Run the tests” sounds harmless until the command installs packages, contacts a service, reads secrets, or runs in the wrong directory. Nerva’s planned terminal workflow should show the exact operation and its boundaries before execution.

A useful approval preview

Command: npm test -- --runInBand worker.test.ts
Directory: /workspace/cryptoex
Network: not requested
Writes: test cache may change
Credentials: not requested
Expected check: worker test result

[Run once] [Deny]

The preview gives the developer enough information to catch a wrong script or path. It also avoids treating an entire terminal session as one permanently approved capability.

Commands change scope quickly

A package script can invoke another command, access the network, or write outside the visible directory. Nerva cannot promise that a command is safe by inspecting only its first word. The tool gateway should capture the working directory, environment policy, and resulting files, while the permission engine applies the user’s rules.

If a test fails because dependencies are missing, Nerva should stop and explain the next proposed action. It should not silently run an install, use administrator privileges, or switch to a remote environment.

The goal is not to make terminal work painful. It is to keep the action legible: a developer can move quickly on routine local checks while seeing exactly when a request becomes a write, network call, or escalation.

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 making command execution narrow and reviewable. 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 previewing a test command with its exact directory, environment, and network behavior. 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 how the system behaves when a command becomes broader after a failure. 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.

The boundary that matters in production

A command preview should include shell interpretation details where they matter: pipes, redirects, substitutions, elevated privileges, and network calls. Approval of npm test is not automatically approval of a later command that installs a package or uploads output. After execution, Nerva should capture exit status and bounded output, then ask for permission again if recovery changes the operation.

For the first vertical slice, keep the boundary small enough to explain in a code review. Define the input, the decision, the side effect, and the evidence for completion. Add a test for the denied path as well as the successful path. This is where a local-first design earns trust: the user can inspect the context and history on the device, see when a provider is involved, and stop before an action crosses the permission boundary. The design should remain honest about its limits. Nerva is planned software, not a claim that every workflow can be made safe automatically. A clear “cannot verify” state is more useful than a polished but unsupported success message.

A review of this design should use a small fixture repository and record both allowed and rejected outcomes. That keeps the discussion concrete: reviewers can compare the requested scope with the selected context, action, and evidence. It also prevents Nerva from being judged only by how persuasive its response sounds.

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.

A final implementation note is to keep this behavior observable in logs and in the interface. A developer should be able to inspect the decision after the task ends, not only while a spinner is running.