CI rehearsal only (do not merge)

[michael-moffett]

Fork CI only. Do not merge.

[greptile-apps]

Confidence Score: 4/5

[Medium risk] Adds test-only cheatcode for minting operations.

The PR appears safe to merge from the reviewed behavior, with a non-blocking instruction-level test gap for cheatcode-created confidential mints.

Findings

1.  Missing instruction-level coverage ▶

Summary

The PR adds surfnet_setMint for creating or updating SPL Token and Token-2022 mints, including a confidential-transfer mint extension, and exposes the method through the Node SDK.

• It preserves existing extension data and tops up rent when the mint grows.
• The new tests do not exercise a real confidential-transfer instruction against a cheatcode-created mint.

Diagram

%%{init: {'theme': 'neutral'}}%%
flowchart TD
  A["surfnet_setMint"] --> B["Load mint or create default"]
  B --> C["Validate owner and apply base fields"]
  C --> D{Confidential update?}
  D -- Yes --> E["Merge Token-2022 mint extension"]
  D -- No --> F["Preserve existing extension bytes"]
  E --> G["Top up rent and write account"]
  F --> G

Loading

_{Reviews (1) · Last reviewed commit: "feat(core): surfnet_setMint cheatcode wi..."}

[greptile-apps]

Missing instruction-level coverage The new test creates a mint with set_mint, then verifies it through set_token_account and get_confidential_balance. Those paths write and read account data directly, so the test does not establish that a Token-2022 confidential-transfer instruction accepts the generated mint or uses its configured auditor. An instruction-level test would catch compatibility problems that this round-trip cannot; the existing instruction-level test creates its mint through Token-2022 instructions instead.

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!