fix: wallet list writes only JSON to stdout (#154)

[ygd58]

Fixes #154.

Problem

nansen wallet list wrote its human-readable summary directly to stdout via console.log and returned undefined, so the CLI's normal {success, data} JSON envelope was never emitted. nansen wallet list | jq . had nothing valid to parse — breaking scripting/agent use, which this CLI is explicitly built for.

Fix

• The list handler now returns the listWallets() result instead of undefined, so the existing CLI dispatcher wraps it in the standard {"success":true,"data":{...}} envelope and prints it to stdout — consistent with every other command.
• The human-readable summary (and the "No wallets found" message) now goes to stderr via the errorOutput dep, so a human running it interactively still gets the friendly view, but stdout stays clean JSON.
• As a side effect, --pretty, --table, and --csv now work on wallet list too, since they're handled generically by formatOutput once the command returns data instead of short-circuiting.
• Updated the existing Privy-provider test to assert against errorOutput instead of log, and added two new tests covering the stdout/stderr split for both the populated and empty-wallet cases.

Testing

npm test

Test Files  1 failed | 65 passed (66)
     Tests  5 failed | 2675 passed | 16 skipped (2696)

The 5 failures are all pre-existing, in src/__tests__/doctor.test.js, and reproduce identically on main without this change — they rely on chmod-based unreadable-file simulation, which doesn't apply when tests run as root. Unrelated to this fix.

npm run lint

Clean, no output.

Manual verification:

$ nansen wallet list | jq .
{
  "success": true,
  "data": { "wallets": [...], "defaultWallet": "..." }
}

Human-readable summary confirmed on stderr, not stdout.

Checklist

• npm test passes (output above; unrelated pre-existing failures noted)
• npm run lint passes
• New code paths have tests
• No console.log in core (uses injected log/errorOutput deps)
• Changeset added (patch — bug fix, user-facing output behavior change)
• src/schema.json — no changes needed (no new commands/options)

[nansen-pr-reviewer]

pr-reviewer Summary for #0b7b1b9

✅ No issues found

The code review completed successfully with no findings.

Review effort: 2/5 (Simple)

Summary

This is a clean, well-scoped bug fix. The core change — having wallet list return listWallets() instead of undefined, and routing the human-readable summary to errorOutput — is exactly the right approach and is consistent with how the rest of the dispatcher works. The fix is minimal, the tests cover both the happy path and the empty-wallet edge case, and the changeset and AGENTS.md are correctly updated.

A few things I verified and found to be fine:

• Dispatcher wiring (cli.js lines 2086–2127): the result === undefined short-circuit fires for operational commands that return nothing; for wallet list the non-undefined return falls through to the formatOutput / JSON envelope path as intended. --pretty, --table, and --csv will now work correctly on this command.
• Test isolation: beforeEach creates a fresh tempDir and stubs HOME, afterEach cleans up. The second new test (no-wallets case) correctly relies on the empty temp dir produced by beforeEach, not on any shared state.
• errorOutput default: falls back to console.error, which is the correct stream for interactive human output and correctly avoids contaminating stdout.
• Changeset: patch semver is appropriate for a bug fix; package name "nansen-cli" matches package.json.
• AGENTS.md update: the exception to the "operational commands return undefined" rule is now correctly documented with the rationale and issue reference.

No actionable findings.

---

Token usage: 4,932 input, 4,362 output, 531,346 cache read, 35,344 cache write | Usage Guide

New pushes are reviewed automatically with a 10-minute cooldown between reviews. To request a review at any time, comment @nansen-pr-reviewer re-review.

[nansen-pr-reviewer]

Auto-approved

This PR was automatically approved because:

• Claude recommends approval
• Claude assessed this as a moderate effort change
• The effort level is within the auto-approval threshold of 2
• No high or critical issues were detected
• Review comment contains non-blocking feedback

If you have any concerns, please request a manual review.

[nansen-pr-reviewer]

Auto-approved

This PR was automatically approved because:

• Claude recommends approval
• Claude assessed this as a moderate effort change
• The effort level is within the auto-approval threshold of 2
• No high or critical issues were detected

If you have any concerns, please request a manual review.