Agent
Evaluates codebases for brittleness, complexity, and coupling. Use when assessing technical debt, reviewing architecture, or planning refactors. Covers dependency analysis and structural issues.
Agent
Evaluates codebases for brittleness, complexity, and coupling. Use when assessing technical debt, reviewing architecture, or planning refactors. Covers dependency analysis and structural issues.
Agent
Designs feature architectures by analyzing codebase patterns. Use when planning new features or major changes. Covers implementation blueprints, component designs, data flows, and build sequences.
Agent
Traces execution paths, maps architecture layers, and documents dependencies. Use when understanding unfamiliar code, tracing features, or mapping data flow. Covers call chains, abstractions, and component relationships.
Agent
Code refinement specialist for improving elegance, readability, and maintainability. Use PROACTIVELY when major developments are completed and you're at a stopping point.
Agent
Reviews code for bugs, logic errors, security vulnerabilities, and convention adherence. Use when reviewing diffs, checking PRs, or validating changes. Covers confidence-based filtering and severity assessment.
Agent
Comment Sicko reviews scoped code comments and suppressions, deleting only comments that fail a strict keep test and reporting structural fixes for the code beneath them.
Agent
Iteratively refine frontend designs until they achieve a 10/10 rating. Use when asked to "refine this design", "polish this UI", "get this to 10/10", or improve visual design quality.
Agent
Documentation maintenance expert for Markdown files, package configs, and developer docs. Use when major changes have been made. Focus on uppercase-named files like README, CONTRIBUTING, CLAUDE, AGENTS, and CHANGELOG.
Agent
Evaluates plans for elegant, maintainable implementations. Suggests alternatives only with high confidence.
Agent
Critically evaluates an individual PR review comment to determine whether any suggested changes should be implemented. Provides reasoned recommendations on whether to fix, skip, or optionally implement suggested changes.
Agent
Adversarial reviewer for investigative conclusions. Spawned after research/analysis to challenge assumptions, identify gaps, and stress-test reasoning before conclusions reach the user.