maji-attack
Performs an adversarial review of work to identify bugs, visual weaknesses, and overhyped claims.
Install
mkdir -p .claude/skills/maji-attack && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/12814" && unzip -o skill.zip -d .claude/skills/maji-attack && rm skill.zipInstalls to .claude/skills/maji-attack
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Adversarially reviews the build for bugs, product risks, and weak claims.Key capabilities
- →Identify bugs in the build
- →Detect product risks
- →Uncover weak claims in the product
- →Critique visual presentation
- →Determine what must be cut from the build
- →Assess what will break in production
How it works
This skill adversarially reviews a build by identifying bugs, product risks, weak claims, and visual weaknesses, then outputs structured JSON with findings.
Inputs & outputs
When to use maji-attack
- →Identify production risks
- →Detect fake or weak claims
- →Perform visual quality assurance
About this skill
You are Maji, the adversary. Attack the work:
- what will break in production?
- what looks fake?
- what is visually weak?
- what claim is overhyped?
- what must be cut? Return structured JSON with:
- status: pass | warn | fail
- issues
- required_fixes
When not to use it
- →When the goal is to fix identified issues directly
- →When the goal is to praise the build
Limitations
- →Only identifies issues, does not fix them
- →Focuses on adversarial review
- →Outputs findings in structured JSON format
How it compares
This workflow adopts an adversarial stance to critically evaluate a build for potential failures and weaknesses, providing a distinct perspective from a standard review.
Compared to similar skills
maji-attack side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| maji-attack (this skill) | 0 | 3mo | No flags | Beginner |
| reviewing-code | 21 | 8mo | No flags | Intermediate |
| pr-review | 6 | 2mo | Review | Intermediate |
| code-review-checklist | 3 | 27d | No flags | Beginner |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
reviewing-code
CaptainCrouton89
Systematically evaluate code changes for security, correctness, performance, and spec alignment. Use when reviewing PRs, assessing code quality, or verifying implementation against requirements.
pr-review
pytorch
Review PyTorch pull requests for code quality, test coverage, security, and backward compatibility. Use when reviewing PRs, when asked to review code changes, or when the user mentions "review PR", "code review", or "check this PR".
code-review-checklist
vudovn
Code review guidelines covering code quality, security, and best practices.
fix-review
sickn33
Verify fix commits address audit findings without new bugs
django-verification
affaan-m
Verification loop for Django projects: migrations, linting, tests with coverage, security scans, and deployment readiness checks before release or PR.
agent-reviewer
ruvnet
Agent skill for reviewer - invoke with $agent-reviewer