share: Show the master fingerprint after the threshold

[BenWestgate]

V1 status: Deferred to post-v1. Current behavior is Codex-reviewed; this enhancement is not required by the supplied adversarial audit and was removed from the frozen-candidate scope after late-stack overlap with #101 was confirmed.

Requested by Ben · project thread

Problem

ms32 secret shows the recovered master fingerprint, but ms32 share did not, so an operator deriving a new card could not compare the recovered seed with the separate wallet record before writing that card.

Change

On an interactive terminal, ms32 share prints Master fingerprint: XXXXXXXX above the derived share whether the threshold was entered as ordinary shares or included S. --plain and redirected output stay unchanged. The fingerprint is derived only after a threshold-capable basis is present.

Review stack

This post-v1 branch currently contains #99 as an ancestor after stack-maintenance PR #123. The unrelated <5250 cap commit was removed and its old tip is preserved on archive/100-pre-restack-20261004.

The focused #100 delta remains the reviewed cli.py / test_cli.py fingerprint-after-threshold change. It is intentionally outside the v1 candidate; rebase it onto the eventual post-v1 runtime before integration.

Validation

The focused behavior has a Codex ACK. Prior validation passed 926 tests plus Ruff and strict mypy. Existing share/recovery validation still requires a threshold-capable compatible basis before a fingerprint can be recovered; redirected and --plain output behavior is covered by tests.

Closes #86.

AI-assisted stack maintenance and review. Responsible-human review/authorship policy still applies before integration.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[BenWestgate]

AI-generated review (Codex), posted at the maintainer's request.

Not ACKing a3a27b0 yet: this raises the installed-code budget from 5,200 to 5,250 without the explicit authorization AGENTS.md requires. The fingerprint-after-threshold behavior itself looks sound.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[BenWestgate]

AI-generated current-head follow-up review (Codex), posted at the maintainer's request.

ACK 8aaf0facce2d now that the unrelated <5250 cap commit has been removed.

The change displays the master fingerprint only on the interactive ms32 share path after a threshold-capable basis is available. --plain and redirected output remain unchanged, and the derivation/recovery validation that establishes the seed is unchanged. No security or correctness blocker found in this focused diff.

The PR is stacked on #105 and preserves <5200; run final CI on the actual integrated #105 + #100 tip before freeze. Responsible-human authorship/review policy still applies.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[BenWestgate]

Codex current-head review at 13fef0f: no findings. Since the prior ACK, the only new commit brings in #105’s already exact-head-reviewed behavior-preserving cleanup; the fingerprint-after-threshold behavior is unchanged and remains confined to interactive output.

[BenWestgate]

AI-generated current-head stack review (Codex), posted at the maintainer's request.

Code ACK 13fef0f39ec9 for the actual #105 + #100 head. #105 is now an ancestor of this branch; comparing #105 (361feb7) to the head yields only the focused cli.py and test_cli.py fingerprint-after-threshold changes. The unrelated <5250 cap change is absent.

The behavior remains scoped to interactive ms32 share: the fingerprint is available only after a threshold-capable basis is recovered, while --plain and redirected output remain unchanged.

Fresh current-head Python-package run #752 is still queued at the time of this review. Treat this as an exact-head code ACK with CI pending. Responsible-human authorship/review still applies before integration.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[BenWestgate]

V1 disposition (Codex, 2026-10-04): defer this post-audit enhancement from the v1 candidate.

The focused behavior is reviewed and the current branch is cleanly stacked after #99, but the feature is not required by any validated DeepSeek/GLM/Kimi/consolidated audit finding. Attempting to carry #100 and #101 together exposed an actual overlap conflict in the late CLI stack. Resolve that after v1 rather than expanding the frozen-candidate review scope.

Keep this PR open for post-v1; its current-head code ACK remains useful. Human authorship/review is still required when integrated.

[BenWestgate]

Codex current-head review at 14b27b2: no findings. The restack now shows only #100’s reviewed fingerprint-after-threshold delta on top of current #99; the runtime/test change is unchanged in substance and still keeps the fingerprint on interactive, non-plain output only.