Productivity

adversarial review - Claude MCP Skill

Adversarial Review Framework

SEO Guide: Enhance your AI agent with the adversarial review tool. This Model Context Protocol (MCP) server allows Claude Desktop and other LLMs to adversarial review framework... Download and configure this skill to unlock new capabilities for your AI workflow.

🌟1 stars • 0 forks
📥0 downloads

Documentation

SKILL.md
# Adversarial Review Framework

**Auto-activate when:** User mentions red-team, adversarial review, find flaws, edge cases, blind spots, attack vectors, "what could go wrong", "poke holes in this", or asks to critique/challenge a plan or design.

**Best for**: Red-team attack on plans/systems to find flaws, edge cases, blind spots

## Template (5 Phases)

### {BASE_TARGET}

#### Phase 1: Challenge Assumptions
- What assumptions does this make?
- Which assumptions are most likely wrong?
- What happens if each assumption fails?

#### Phase 2: Edge Case Mining
- Boundary conditions (empty, null, max, negative)
- Timing issues (race conditions, ordering)
- Environment variations (OS, versions, permissions)
- Data quality issues (malformed, missing, duplicate)

#### Phase 3: Failure Mode Analysis
- List 5 ways this could fail
- For each: Likelihood (H/M/L), Impact (1-10), Mitigation
- Which failures are catastrophic?

#### Phase 4: Hidden Dependencies
- Undocumented state assumptions
- External service dependencies
- Implicit ordering requirements
- Configuration assumptions

#### Phase 5: Attack Vectors & Blind Spots
- What did we not consider?
- What expertise are we missing?
- Where could malicious input cause issues?
- What would break this in production?

## Principles

Be adversarial. If you don't find issues, you weren't critical enough.

Signals

Avg rating0.0
Reviews0
Favorites0

Information

Repository
ilude/claude-code-config
Author
ilude
Last Sync
9/4/2026
Repo Updated
7/16/2026
Created
1/15/2026

Reviews (0)

No reviews yet. Be the first to review this skill!

Related Skills

code-review

Perform thorough code reviews with security, performance, and maintainability analysis. Use when user asks to review code, check for bugs, or audit a codebase.

19016

claude-api

Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).

8706

update-chrome-binaries-test-region

- Confirm the AWS CLI is logged into account `678892195805` by running `aws sts get-caller-identity`. If not, ask the user to log in.

6396

copilotkit-setup

Use when adding CopilotKit to an existing project or bootstrapping a new CopilotKit project from scratch. Covers framework detection, package installation, runtime wiring (managed Intelligence or self-hosted SSE), provider setup, and first working chat integration.

1618

Related Guides