robertoecf/adversarial-review

3 stars · Last commit 2026-04-28

Review triad plugin for Claude Code: prompt optimization, adversarial red-team review, plan validation, and intelligent orchestration.

README preview

# adversarial-review

Cross-host adversarial review for coding workflows. Works in **both Claude
Code and Codex** from the same source — detects which agent host you're
running in and routes the heavy critique to the OTHER agent.

## The principle

> **The partner reviews, never the host.**

Two models examining the same artifact from different angles catch more
issues than either alone. Each has different training biases, blind spots,
and reasoning patterns. Disagreements between the two surface the
highest-value findings — the ones a single reviewer would miss.

This plugin enforces that principle automatically:

- Running in **Claude Code** → external reviewer is **Codex** (`codex exec`,
  `gpt-5.4 xhigh` via ChatGPT subscription auth)
- Running in **Codex** → external reviewer is **Claude (Opus, xhigh)** via

View full repository on GitHub →