Whos there

[wrjones104]

Feature: Who's There?

Summary

Implements the -who flag, which replaces all boss sprites with the standard Imp sprite and replaces their display names with ??????.

Technical Implementation

1. Name Obfuscation

• nIterates through the bosses.enemy_name index mapping.
• Replaces boss name strings in the ROM with ?????? (hex 0xBF).

2. Assembly Graphics Hook

• Injects a custom 65816 assembly subroutine into Bank C0.
• Patches the battle graphics loading routine at ROM offset 0x01207B (Bank C1).
• Logic: The routine reads the 16-bit monster ID from WRAM $2001,X (indexed by slot) and compares it against a 384-byte boss_table in Bank F0. If a match is found, it forces the graphics pointer to the Imp index (#$0816).

3. VRAM/Tilemap Corruption Fix

• Issue: Large or multi-part bosses (e.g., Tentacles, Whelk) use custom tilemap dimensions. Forcing the small Imp sprite into these large maps causes PPU read errors and VRAM corruption.
• Solution: Within the .is_imp branch, the routine overwrites the active graphics index array ($812F,X) with 0 (Guard ID). This forces the dimension calculator (JSR $202F) to assign standard 32x32 properties, ensuring multi-part components render as individual, stable Imps.

Exceptions & Edge Cases

• Included: SrBehemoth (ID 282) was manually added to the lookup table to ensure both phases are transformed. Final Kefka (ID 298) is also transformed.
• Excluded: Final Battle Tiers (IDs 343–351) to preserve end-game aesthetics.
  Phantom Train (rendered as a background layer rather than a sprite object).

[gemini-code-assist]

Code Review

This pull request introduces the "Who's There?" feature, which visually transforms bosses into Imps and obfuscates their names to "??????". The changes span command-line argument configuration, documentation updates, and a custom assembly subroutine in data/enemies.py to hook the graphics loader. The review feedback highlights several critical issues in the assembly generation code, including an IndentationError from duplicated loop headers, potential CPU register corruption due to unpreserved Y register modifications, and multiple instances of duplicated code, imports, and ROM space allocations.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[wrjones104]

/gemini review

[wrjones104]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces two new features: 'Who's There?' (--who-there), which obfuscates boss names and forces them to render as Imps, and 'Steveify' (--steveify), which replaces various in-game names with a custom string. It also updates documentation regarding assembly hooks and register preservation. However, several critical issues must be addressed: the custom assembly subroutine in data/enemies.py contains bugs that lead to register corruption and an overwritten return value, the graphics loader patch leaves garbage bytes that will cause a crash, and the --steveify command-line argument is missing from the parser in args/graphics.py.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

Code Review

This pull request introduces the 'Who's There?' feature, which makes bosses look like Imps and obscures their names, and the 'Steveify' feature, which allows renaming various game elements to a custom string. However, several critical issues were identified during the review: the command-line argument for 'Steveify' is missing from the parser, which will cause an immediate startup crash; the custom assembly subroutine for 'Who's There?' contains register corruption, a truncation bug for 16-bit monster IDs, and a processor status register bug; and the graphics loader patch overwrites too many bytes, corrupting the instruction stream and using an incorrect branch offset.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[wrjones104]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces the "Who's There?" (-who / --who-there) option, which obfuscates boss names to "??????" and dynamically renders them as Imps. This is achieved via a custom assembly subroutine in Bank C0 hooked into the graphics loader at ROM offset 0x01207B. The documentation in agents.md and llms.md has been updated to reflect assembly hook guidelines and implementation details. Feedback on the changes highlights a critical bug where the graphics loader patch only overwrites 9 bytes instead of 10, leaving a stray byte that corrupts the instruction stream; adding a fourth NOP is recommended. Additionally, a redundant block of code for parsing and validating character names in args/graphics.py should be removed.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.