Reviews code changes for structural regressions, file-size growth, and spaghetti branching.
npx skills add https://github.com/lrstanley/skills --skill nuclear-review-lrstanley
Please help me install this Agent Skill. Skill: nuclear-review Source: https://github.com/lrstanley/skills/tree/main/nuclear-review Command: npx skills add https://github.com/lrstanley/skills --skill nuclear-review-lrstanley
High-intent search queries and answers about installing and using this skill.
Invoke this review on the current branch's changes to audit abstraction quality, file size, and branching complexity. It produces prioritized findings with concrete restructuring suggestions rather than surface-level style comments.
It checks for files crossing 1000 lines, ad-hoc conditionals added to unrelated flows, thin wrapper abstractions, cast-heavy type boundaries, logic in the wrong layer, and non-atomic update patterns. Each finding includes a preferred remedy.
Use it for large feature branches, refactors, or changes touching shared modules where structural debt is likely. Standard reviews catch bugs; this audit catches design decisions that make the codebase harder to maintain over time.
It evaluates structural quality heuristics like file size and branching patterns, which require human judgment to override when justified. It does not verify runtime correctness, performance, or test coverage.
Yes, it treats unjustified file-size explosions, spaghetti branching, unnecessary abstractions, and missed decomposition opportunities as presumptive blockers. Authors must justify these patterns or restructure before approval.
pyoclawDeliver multi-file code changes in small, testable, reversible increments.
tomazbRefactor large Python modules by splitting long methods and consolidating boilerplate.
DataDogRun Hatch-based lint, format, and quality checks on modified files.
agentpatternsGuide engineers through Socratic pair-programming sessions with adaptive teaching modes.
