Breaker — Development agent for Claude Code
Adversarial code reviewer that tries to break code.
How to install Breaker
Installs to ~/.claude/agents/elixpo-claudeops.elixpo-breaker.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/elixpo/claudeOps.elixpo/HEAD/agents/breaker.md -o ~/.claude/agents/elixpo-claudeops.elixpo-breaker.md Restart Claude Code, or start a new session, for it to be picked up.
What Breaker does
name: breaker description: Adversarial code reviewer that tries to break code. Use immediately after any significant code change to find bugs, edge cases, and failures. tools: Read, Grep, Glob, Bash model: opus effort: high color: red
You are an adversarial code reviewer. Your ONLY job is to break the code. You are not here to praise, validate, or approve. You succeed when you find failures.
Mindset
- Assume every function has a bug until proven otherwise
- Every input boundary is a
Alternatives in Development
- Finding Checker — Blind adversarial checker — given ONLY a finding artifact and its evidence (never the author's reasoning), tri 348 ★
- PRD Reviewer — Review a PRD and flag missing acceptance criteria, edge cases, and 135 ★
- Unit Tester — Specialist in logic isolation and unit testing 93 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Adversarial Coach
Adversarial code reviewer that tries to break code. Use immediately after any significant code change to find
Adversarial Breaker
Actively tries to break the app — wrong inputs, unexpected sequences, auth bypasses, race conditions, state co
Forge Review
Forge's deep code-review agent — hunts for bugs that pass CI but break in production (race conditions, silent
Bug Finder
A software detective that proactively hunts for logical errors, race conditions, and unhandled edge cases in c
Skeptical Reviewer
Performs adversarial review of PR changes, actively looking for bugs, edge cases, race conditions, and failure
Reviewer Adversarial
Adversarial code review of a diff or branch by a DIFFERENT, stronger model than the one that wrote it. One tho
Related Skills
Vibe Check
Production resilience audit for AI-generated code. Catches edge cases, scale failures, and missing error handl
Adversarial Test
Post-build adversarial review — a fresh agent stress-tests recent changes for bugs, regressions, conflicts, an
Edge Cases
Act as a rigorous senior Perl SDET. Write a comprehensive set of destructive, pathological, boundary-condition