Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
alirezarezvani avatar

Adversarial Reviewer

  • 646 installs
  • 23.5k repo stars
  • Updated July 17, 2026
  • alirezarezvani/claude-skills

adversarial-reviewer is a Claude agent skill that runs a hostile, persona-driven review on recent diffs when normal self-review feels too agreeable for developers preparing to merge code.

About

adversarial-reviewer is an alirezarezvani/claude-skills skill that deliberately attacks recent code diffs from adversarial personas instead of offering agreeable self-review. The skill targets the window before merge when teams want dissenting voices to surface edge cases, security gaps, logic holes, and unstated assumptions in a pull request or local diff. Developers reach for adversarial-reviewer when a change feels rubber-stamped or when standard review prompts return overly polite feedback. The workflow focuses on recent diffs so critique stays anchored to concrete line-level changes rather than abstract architecture lectures. adversarial-reviewer complements conventional review by forcing explicit rebuttal or fixes before code lands on main.

  • Three adversarial personas: Saboteur, New Hire, and Security Auditor (OWASP-informed)
  • Mandatory findings — each persona must surface at least one issue (no rubber-stamp LGTM)
  • Severity promotion when two or more personas agree on the same issue
  • Structured verdicts: BLOCK, CONCERNS, or CLEAN with merge guidance
  • Self-review trap breaker techniques for shared author–reviewer mental models

Adversarial Reviewer by the numbers

  • 646 all-time installs (skills.sh)
  • Ranked #206 of 1,356 Code Review & Quality skills by installs in the Skillselion catalog
  • Security screen: CRITICAL risk (skills.sh audit)
  • Data as of Jul 31, 2026 (Skillselion catalog sync)
npx skills add https://github.com/alirezarezvani/claude-skills --skill adversarial-reviewer

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs646
repo stars23.5k
Security audit1 / 3 scanners passed
Last updatedJuly 17, 2026
Repositoryalirezarezvani/claude-skills

How do you stress-test a diff before merging?

Run a hostile, persona-driven review on recent diffs when normal self-review feels too agreeable before you merge.

Who is it for?

Developers about to merge a non-trivial diff who want deliberately hostile, persona-driven critique beyond polite self-review.

Skip if: Developers who need automated lint fixes, security CVE scanning, or greenfield architecture planning without an existing diff.

When should I use this skill?

The user wants a hostile or adversarial review of recent diffs before merge because normal review feels too agreeable.

What you get

Persona-driven adversarial review notes listing risks, gaps, and challenged assumptions in the diff.

  • adversarial review report
  • challenged-assumption list

Files

SKILL.mdMarkdownGitHub ↗

../../../engineering-team/skills/adversarial-reviewer/SKILL.md

Related skills

How it compares

Use adversarial-reviewer for dissenting diff critique before merge; use sparkfinderoven-security-compliance-skills when the need is OWASP scans or compliance audits.

FAQ

What does adversarial-reviewer review?

adversarial-reviewer focuses on recent code diffs—typically a pull request or local changes—applying hostile persona voices to challenge logic, edge cases, and unstated assumptions before merge.

When should adversarial-reviewer run?

adversarial-reviewer should run during the pre-merge ship phase when standard or self-review feels too agreeable and the team wants deliberately dissenting critique on the actual diff.

How is adversarial-reviewer different from normal review?

adversarial-reviewer uses persona-driven hostile questioning to attack diffs rather than summarize politely, surfacing risks that agreeable review prompts often skip before code merges.

Is Adversarial Reviewer safe to install?

skills.sh reports 1 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.