neurofoo/redteam
Red team adversarial analysis to find weaknesses, vulnerabilities, and failure modes. Use before launches, for security review, or when a plan feels too perfect.
npx skills add https://github.com/neurofoo/agent-skills --skill redteam
Conduct an adversarial analysis to find weaknesses, vulnerabilities, and failure modes in a plan, system, or strategy.
Think like an attacker. Your job is to break this, find the gaps, and expose what could go wrong. Be thorough and creative in your attacks.
Target: [What we're attacking]
Objective: [What would "breaking it" look like?]
Summary of the Plan/System
[Brief description of what we're analyzing]
Key Assumptions
Entry Points / Vulnerabilities
| Vector | Description | Severity |
|--------|-------------|----------|
| [attack vector] | [how it could be exploited] | Critical/High/Med/Low |
| Failure Mode | Trigger | Impact |
|--------------|---------|--------|
| [what could fail] | [what causes it] | [effect] |
| Failure Mode | Trigger | Impact |
|--------------|---------|--------|
| [human error] | [situation] | [consequence] |
If I wanted this to fail, I would...
| Assumption | Attack | What If Wrong? |
|------------|--------|----------------|
| [assumption] | [challenge to it] | [consequences] |
What are we not seeing?
What are we too optimistic about?
| Vulnerability | Risk | Mitigation |
|---------------|------|-----------|
| [vulnerability] | [risk level] | [how to fix] |
| Concern | Recommendation |
|---------|----------------|
| [concern] | [recommendation] |
Immediate actions:
Ongoing monitoring:
Bottom Line
> [Is this plan/system ready? What must change?]
$ARGUMENTS
Take neurofoo/redteam from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.