# Adversarial Review

> Cross-model code review. Sends code to a DIFFERENT model (GPT/Gemini) for independent review. Reviewer must be on a different model than the code author. Use for critical PRs and security-sensitive code.

- **Type:** Skill
- **Install:** `agentstack add skill-neuron-one-godmode-adversarial-review`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [neuron-one](https://agentstack.voostack.com/s/neuron-one)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [neuron-one](https://github.com/neuron-one)
- **Source:** https://github.com/neuron-one/GODMODE/tree/master/skills/multimodel/adversarial-review

## Install

```sh
agentstack add skill-neuron-one-godmode-adversarial-review
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Adversarial Cross-Model Review

## Principle
Same model has same blind spots when writing AND reviewing.
Different model catches what the author's model missed.

## Process
1. Identify code changes for review
2. Select DIFFERENT model than what wrote the code
3. Send code + context to external model
4. Collect review findings
5. Compare with internal review
6. Address all unique findings from external model

## Rules
- Reviewer MUST be different model
- Claude wrote code -> GPT reviews (or vice versa)
- Include sufficient context for reviewer
- Don't dismiss external findings without investigation

## Source & license

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

- **Author:** [neuron-one](https://github.com/neuron-one)
- **Source:** [neuron-one/GODMODE](https://github.com/neuron-one/GODMODE)
- **License:** MIT

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

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-neuron-one-godmode-adversarial-review
- Seller: https://agentstack.voostack.com/s/neuron-one
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.