AgentStack
SKILL verified MIT Self-run

Redteam

skill-neurofoo-agent-skills-redteam · by neurofoo

Red team adversarial analysis to find weaknesses, vulnerabilities, and failure modes. Use before launches, for security review, or when a plan feels too perfect.

No reviews yet
0 installs
13 views
0.0% view→install

Install

$ agentstack add skill-neurofoo-agent-skills-redteam

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Redteam? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Red Team Analysis

Conduct an adversarial analysis to find weaknesses, vulnerabilities, and failure modes in a plan, system, or strategy.

Instructions

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.

Output Format

Target: [What we're attacking] Objective: [What would "breaking it" look like?]


Target Understanding

Summary of the Plan/System [Brief description of what we're analyzing]

Key Assumptions

  • [Assumption 1]
  • [Assumption 2]

Attack Surface Analysis

Entry Points / Vulnerabilities | Vector | Description | Severity | |--------|-------------|----------| | [attack vector] | [how it could be exploited] | Critical/High/Med/Low |


Failure Mode Analysis

Technical/Operational Failures

| Failure Mode | Trigger | Impact | |--------------|---------|--------| | [what could fail] | [what causes it] | [effect] |

Human Failures

| Failure Mode | Trigger | Impact | |--------------|---------|--------| | [human error] | [situation] | [consequence] |


Adversary Scenarios

If I wanted this to fail, I would...

Scenario 1: [Attack Name]

  • Attack method: [how they'd do it]
  • Likelihood of success: [High/Med/Low]
  • Impact if successful: [consequences]

Scenario 2: [Attack Name]

  • Attack method: [how]
  • Likelihood of success: [High/Med/Low]
  • Impact if successful: [consequences]

Assumption Attacks

| Assumption | Attack | What If Wrong? | |------------|--------|----------------| | [assumption] | [challenge to it] | [consequences] |


Blind Spot Analysis

What are we not seeing?

  • [Blind spot 1]
  • [Blind spot 2]

What are we too optimistic about?

  • [Over-optimism 1]

Red Team Findings

Critical Vulnerabilities (Must Address)

| Vulnerability | Risk | Mitigation | |---------------|------|-----------| | [vulnerability] | [risk level] | [how to fix] |

High-Priority Concerns

| Concern | Recommendation | |---------|----------------| | [concern] | [recommendation] |


Hardening Recommendations

Immediate actions:

  1. [Action 1]
  2. [Action 2]

Ongoing monitoring:

  1. [What to watch]

Bottom Line > [Is this plan/system ready? What must change?]

Guidelines

  • Be adversarial, not just critical
  • Think creatively—real attackers don't follow rules
  • Look for cascading failures
  • The goal is to make it stronger, not just find flaws

$ARGUMENTS

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.