sherlock-review
About
Sherlock-review is an evidence-based code review skill that uses deductive reasoning to systematically verify implementation claims, investigate bugs, and perform root cause analysis. It guides developers through observing concrete evidence and deducing what actually happened versus what was stated. This makes it ideal for validating fixes, security audits, and performance validation during code reviews.
Quick Install
Claude Code
Recommendednpx skills add proffesor-for-testing/agentic-qe -a claude-code/plugin add https://github.com/proffesor-for-testing/agentic-qegit clone https://github.com/proffesor-for-testing/agentic-qe.git ~/.claude/skills/sherlock-reviewCopy and paste this command in Claude Code to install this skill
GitHub Repository
Related Skills
sherlock-review
OtherSherlock Review performs evidence-based code investigation using deductive reasoning to verify implementation claims and find root causes. It systematically analyzes code, tests, and history to determine what actually happened versus what was reported. Use this skill for validating fixes, investigating bugs, or conducting rigorous code reviews.
brutal-honesty-review
OtherThis skill provides direct, unfiltered technical criticism for code reviews, testing strategies, or technical decisions that require a harsh reality check. It operates in three distinct modes—Linus (precision), Ramsay (standards), and Bach (BS detection)—to surgically identify flaws without sugar-coating. Use it when you need a brutally honest assessment of what's broken and why.
brutal-honesty-review
OtherThis skill provides direct, unfiltered technical criticism for code reviews, testing, or architectural decisions by channeling distinct analytical modes. It surgically identifies flaws and their root causes without sugar-coating, aimed at situations requiring harsh reality checks. Developers should use it when they need rigorous, no-nonsense feedback to expose broken logic or insufficient rigor.
brutal-honesty-review
OtherThis skill provides direct, unfiltered technical criticism for code reviews, testing, or architectural decisions, modeled after the precision of expert critics. It operates in three distinct modes—technical accuracy, quality standards, and logic validation—to deliver surgical feedback without sugar-coating. Use it when you need a harsh reality check to identify exactly what is broken and why.
