| name | deep-review |
| description | Runs an aggressively strict maintainability review focused on abstraction and complexity risk. Use when a harsh deep quality audit is requested. |
| disable-model-invocation | true |
Deep Review
Run an aggressive maintainability review focused on structural simplification.
Review Priorities
- Prefer designs that delete complexity, not move it.
- Treat new spaghetti branching and ad-hoc flags as design problems.
- Flag file-size explosions (especially crossing ~1000 lines) unless justified.
- Push logic toward the appropriate layer and reuse existing helpers.
- Prefer explicit type/boundary contracts over cast-heavy or optionality-heavy flow.
- Call out unnecessary indirection, wrappers, and magic abstractions.
- Prefer simpler, more atomic orchestration when obvious.
Core Questions
- Is there a clear "code judo" simplification that removes branches/helpers/layers?
- Did this change make the module easier or harder to reason about?
- Is logic placed in the right layer with the right abstraction boundary?
- Did the diff add avoidable conditionals, casts, or duplicated logic?
- Is decomposition needed before accepting the current shape?
Output Expectations
- Lead with highest-severity structural findings.
- Use concrete evidence and actionable remedies.
- Prefer a few high-conviction findings over many cosmetic notes.
- Treat major structural regressions as blockers unless justified.
